Annotation of imach/src/imach.c, revision 1.241
1.241 ! brouard 1: /* $Id: imach.c,v 1.240 2016/08/29 07:53:18 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.241 ! brouard 4: Revision 1.240 2016/08/29 07:53:18 brouard
! 5: Summary: Better
! 6:
1.240 brouard 7: Revision 1.239 2016/08/26 15:51:03 brouard
8: Summary: Improvement in Powell output in order to copy and paste
9:
10: Author:
11:
1.239 brouard 12: Revision 1.238 2016/08/26 14:23:35 brouard
13: Summary: Starting tests of 0.99
14:
1.238 brouard 15: Revision 1.237 2016/08/26 09:20:19 brouard
16: Summary: to valgrind
17:
1.237 brouard 18: Revision 1.236 2016/08/25 10:50:18 brouard
19: *** empty log message ***
20:
1.236 brouard 21: Revision 1.235 2016/08/25 06:59:23 brouard
22: *** empty log message ***
23:
1.235 brouard 24: Revision 1.234 2016/08/23 16:51:20 brouard
25: *** empty log message ***
26:
1.234 brouard 27: Revision 1.233 2016/08/23 07:40:50 brouard
28: Summary: not working
29:
1.233 brouard 30: Revision 1.232 2016/08/22 14:20:21 brouard
31: Summary: not working
32:
1.232 brouard 33: Revision 1.231 2016/08/22 07:17:15 brouard
34: Summary: not working
35:
1.231 brouard 36: Revision 1.230 2016/08/22 06:55:53 brouard
37: Summary: Not working
38:
1.230 brouard 39: Revision 1.229 2016/07/23 09:45:53 brouard
40: Summary: Completing for func too
41:
1.229 brouard 42: Revision 1.228 2016/07/22 17:45:30 brouard
43: Summary: Fixing some arrays, still debugging
44:
1.227 brouard 45: Revision 1.226 2016/07/12 18:42:34 brouard
46: Summary: temp
47:
1.226 brouard 48: Revision 1.225 2016/07/12 08:40:03 brouard
49: Summary: saving but not running
50:
1.225 brouard 51: Revision 1.224 2016/07/01 13:16:01 brouard
52: Summary: Fixes
53:
1.224 brouard 54: Revision 1.223 2016/02/19 09:23:35 brouard
55: Summary: temporary
56:
1.223 brouard 57: Revision 1.222 2016/02/17 08:14:50 brouard
58: Summary: Probably last 0.98 stable version 0.98r6
59:
1.222 brouard 60: Revision 1.221 2016/02/15 23:35:36 brouard
61: Summary: minor bug
62:
1.220 brouard 63: Revision 1.219 2016/02/15 00:48:12 brouard
64: *** empty log message ***
65:
1.219 brouard 66: Revision 1.218 2016/02/12 11:29:23 brouard
67: Summary: 0.99 Back projections
68:
1.218 brouard 69: Revision 1.217 2015/12/23 17:18:31 brouard
70: Summary: Experimental backcast
71:
1.217 brouard 72: Revision 1.216 2015/12/18 17:32:11 brouard
73: Summary: 0.98r4 Warning and status=-2
74:
75: Version 0.98r4 is now:
76: - displaying an error when status is -1, date of interview unknown and date of death known;
77: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
78: Older changes concerning s=-2, dating from 2005 have been supersed.
79:
1.216 brouard 80: Revision 1.215 2015/12/16 08:52:24 brouard
81: Summary: 0.98r4 working
82:
1.215 brouard 83: Revision 1.214 2015/12/16 06:57:54 brouard
84: Summary: temporary not working
85:
1.214 brouard 86: Revision 1.213 2015/12/11 18:22:17 brouard
87: Summary: 0.98r4
88:
1.213 brouard 89: Revision 1.212 2015/11/21 12:47:24 brouard
90: Summary: minor typo
91:
1.212 brouard 92: Revision 1.211 2015/11/21 12:41:11 brouard
93: Summary: 0.98r3 with some graph of projected cross-sectional
94:
95: Author: Nicolas Brouard
96:
1.211 brouard 97: Revision 1.210 2015/11/18 17:41:20 brouard
98: Summary: Start working on projected prevalences
99:
1.210 brouard 100: Revision 1.209 2015/11/17 22:12:03 brouard
101: Summary: Adding ftolpl parameter
102: Author: N Brouard
103:
104: We had difficulties to get smoothed confidence intervals. It was due
105: to the period prevalence which wasn't computed accurately. The inner
106: parameter ftolpl is now an outer parameter of the .imach parameter
107: file after estepm. If ftolpl is small 1.e-4 and estepm too,
108: computation are long.
109:
1.209 brouard 110: Revision 1.208 2015/11/17 14:31:57 brouard
111: Summary: temporary
112:
1.208 brouard 113: Revision 1.207 2015/10/27 17:36:57 brouard
114: *** empty log message ***
115:
1.207 brouard 116: Revision 1.206 2015/10/24 07:14:11 brouard
117: *** empty log message ***
118:
1.206 brouard 119: Revision 1.205 2015/10/23 15:50:53 brouard
120: Summary: 0.98r3 some clarification for graphs on likelihood contributions
121:
1.205 brouard 122: Revision 1.204 2015/10/01 16:20:26 brouard
123: Summary: Some new graphs of contribution to likelihood
124:
1.204 brouard 125: Revision 1.203 2015/09/30 17:45:14 brouard
126: Summary: looking at better estimation of the hessian
127:
128: Also a better criteria for convergence to the period prevalence And
129: therefore adding the number of years needed to converge. (The
130: prevalence in any alive state shold sum to one
131:
1.203 brouard 132: Revision 1.202 2015/09/22 19:45:16 brouard
133: Summary: Adding some overall graph on contribution to likelihood. Might change
134:
1.202 brouard 135: Revision 1.201 2015/09/15 17:34:58 brouard
136: Summary: 0.98r0
137:
138: - Some new graphs like suvival functions
139: - Some bugs fixed like model=1+age+V2.
140:
1.201 brouard 141: Revision 1.200 2015/09/09 16:53:55 brouard
142: Summary: Big bug thanks to Flavia
143:
144: Even model=1+age+V2. did not work anymore
145:
1.200 brouard 146: Revision 1.199 2015/09/07 14:09:23 brouard
147: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
148:
1.199 brouard 149: Revision 1.198 2015/09/03 07:14:39 brouard
150: Summary: 0.98q5 Flavia
151:
1.198 brouard 152: Revision 1.197 2015/09/01 18:24:39 brouard
153: *** empty log message ***
154:
1.197 brouard 155: Revision 1.196 2015/08/18 23:17:52 brouard
156: Summary: 0.98q5
157:
1.196 brouard 158: Revision 1.195 2015/08/18 16:28:39 brouard
159: Summary: Adding a hack for testing purpose
160:
161: After reading the title, ftol and model lines, if the comment line has
162: a q, starting with #q, the answer at the end of the run is quit. It
163: permits to run test files in batch with ctest. The former workaround was
164: $ echo q | imach foo.imach
165:
1.195 brouard 166: Revision 1.194 2015/08/18 13:32:00 brouard
167: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
168:
1.194 brouard 169: Revision 1.193 2015/08/04 07:17:42 brouard
170: Summary: 0.98q4
171:
1.193 brouard 172: Revision 1.192 2015/07/16 16:49:02 brouard
173: Summary: Fixing some outputs
174:
1.192 brouard 175: Revision 1.191 2015/07/14 10:00:33 brouard
176: Summary: Some fixes
177:
1.191 brouard 178: Revision 1.190 2015/05/05 08:51:13 brouard
179: Summary: Adding digits in output parameters (7 digits instead of 6)
180:
181: Fix 1+age+.
182:
1.190 brouard 183: Revision 1.189 2015/04/30 14:45:16 brouard
184: Summary: 0.98q2
185:
1.189 brouard 186: Revision 1.188 2015/04/30 08:27:53 brouard
187: *** empty log message ***
188:
1.188 brouard 189: Revision 1.187 2015/04/29 09:11:15 brouard
190: *** empty log message ***
191:
1.187 brouard 192: Revision 1.186 2015/04/23 12:01:52 brouard
193: Summary: V1*age is working now, version 0.98q1
194:
195: Some codes had been disabled in order to simplify and Vn*age was
196: working in the optimization phase, ie, giving correct MLE parameters,
197: but, as usual, outputs were not correct and program core dumped.
198:
1.186 brouard 199: Revision 1.185 2015/03/11 13:26:42 brouard
200: Summary: Inclusion of compile and links command line for Intel Compiler
201:
1.185 brouard 202: Revision 1.184 2015/03/11 11:52:39 brouard
203: Summary: Back from Windows 8. Intel Compiler
204:
1.184 brouard 205: Revision 1.183 2015/03/10 20:34:32 brouard
206: Summary: 0.98q0, trying with directest, mnbrak fixed
207:
208: We use directest instead of original Powell test; probably no
209: incidence on the results, but better justifications;
210: We fixed Numerical Recipes mnbrak routine which was wrong and gave
211: wrong results.
212:
1.183 brouard 213: Revision 1.182 2015/02/12 08:19:57 brouard
214: Summary: Trying to keep directest which seems simpler and more general
215: Author: Nicolas Brouard
216:
1.182 brouard 217: Revision 1.181 2015/02/11 23:22:24 brouard
218: Summary: Comments on Powell added
219:
220: Author:
221:
1.181 brouard 222: Revision 1.180 2015/02/11 17:33:45 brouard
223: Summary: Finishing move from main to function (hpijx and prevalence_limit)
224:
1.180 brouard 225: Revision 1.179 2015/01/04 09:57:06 brouard
226: Summary: back to OS/X
227:
1.179 brouard 228: Revision 1.178 2015/01/04 09:35:48 brouard
229: *** empty log message ***
230:
1.178 brouard 231: Revision 1.177 2015/01/03 18:40:56 brouard
232: Summary: Still testing ilc32 on OSX
233:
1.177 brouard 234: Revision 1.176 2015/01/03 16:45:04 brouard
235: *** empty log message ***
236:
1.176 brouard 237: Revision 1.175 2015/01/03 16:33:42 brouard
238: *** empty log message ***
239:
1.175 brouard 240: Revision 1.174 2015/01/03 16:15:49 brouard
241: Summary: Still in cross-compilation
242:
1.174 brouard 243: Revision 1.173 2015/01/03 12:06:26 brouard
244: Summary: trying to detect cross-compilation
245:
1.173 brouard 246: Revision 1.172 2014/12/27 12:07:47 brouard
247: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
248:
1.172 brouard 249: Revision 1.171 2014/12/23 13:26:59 brouard
250: Summary: Back from Visual C
251:
252: Still problem with utsname.h on Windows
253:
1.171 brouard 254: Revision 1.170 2014/12/23 11:17:12 brouard
255: Summary: Cleaning some \%% back to %%
256:
257: The escape was mandatory for a specific compiler (which one?), but too many warnings.
258:
1.170 brouard 259: Revision 1.169 2014/12/22 23:08:31 brouard
260: Summary: 0.98p
261:
262: Outputs some informations on compiler used, OS etc. Testing on different platforms.
263:
1.169 brouard 264: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 265: Summary: update
1.169 brouard 266:
1.168 brouard 267: Revision 1.167 2014/12/22 13:50:56 brouard
268: Summary: Testing uname and compiler version and if compiled 32 or 64
269:
270: Testing on Linux 64
271:
1.167 brouard 272: Revision 1.166 2014/12/22 11:40:47 brouard
273: *** empty log message ***
274:
1.166 brouard 275: Revision 1.165 2014/12/16 11:20:36 brouard
276: Summary: After compiling on Visual C
277:
278: * imach.c (Module): Merging 1.61 to 1.162
279:
1.165 brouard 280: Revision 1.164 2014/12/16 10:52:11 brouard
281: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
282:
283: * imach.c (Module): Merging 1.61 to 1.162
284:
1.164 brouard 285: Revision 1.163 2014/12/16 10:30:11 brouard
286: * imach.c (Module): Merging 1.61 to 1.162
287:
1.163 brouard 288: Revision 1.162 2014/09/25 11:43:39 brouard
289: Summary: temporary backup 0.99!
290:
1.162 brouard 291: Revision 1.1 2014/09/16 11:06:58 brouard
292: Summary: With some code (wrong) for nlopt
293:
294: Author:
295:
296: Revision 1.161 2014/09/15 20:41:41 brouard
297: Summary: Problem with macro SQR on Intel compiler
298:
1.161 brouard 299: Revision 1.160 2014/09/02 09:24:05 brouard
300: *** empty log message ***
301:
1.160 brouard 302: Revision 1.159 2014/09/01 10:34:10 brouard
303: Summary: WIN32
304: Author: Brouard
305:
1.159 brouard 306: Revision 1.158 2014/08/27 17:11:51 brouard
307: *** empty log message ***
308:
1.158 brouard 309: Revision 1.157 2014/08/27 16:26:55 brouard
310: Summary: Preparing windows Visual studio version
311: Author: Brouard
312:
313: In order to compile on Visual studio, time.h is now correct and time_t
314: and tm struct should be used. difftime should be used but sometimes I
315: just make the differences in raw time format (time(&now).
316: Trying to suppress #ifdef LINUX
317: Add xdg-open for __linux in order to open default browser.
318:
1.157 brouard 319: Revision 1.156 2014/08/25 20:10:10 brouard
320: *** empty log message ***
321:
1.156 brouard 322: Revision 1.155 2014/08/25 18:32:34 brouard
323: Summary: New compile, minor changes
324: Author: Brouard
325:
1.155 brouard 326: Revision 1.154 2014/06/20 17:32:08 brouard
327: Summary: Outputs now all graphs of convergence to period prevalence
328:
1.154 brouard 329: Revision 1.153 2014/06/20 16:45:46 brouard
330: Summary: If 3 live state, convergence to period prevalence on same graph
331: Author: Brouard
332:
1.153 brouard 333: Revision 1.152 2014/06/18 17:54:09 brouard
334: Summary: open browser, use gnuplot on same dir than imach if not found in the path
335:
1.152 brouard 336: Revision 1.151 2014/06/18 16:43:30 brouard
337: *** empty log message ***
338:
1.151 brouard 339: Revision 1.150 2014/06/18 16:42:35 brouard
340: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
341: Author: brouard
342:
1.150 brouard 343: Revision 1.149 2014/06/18 15:51:14 brouard
344: Summary: Some fixes in parameter files errors
345: Author: Nicolas Brouard
346:
1.149 brouard 347: Revision 1.148 2014/06/17 17:38:48 brouard
348: Summary: Nothing new
349: Author: Brouard
350:
351: Just a new packaging for OS/X version 0.98nS
352:
1.148 brouard 353: Revision 1.147 2014/06/16 10:33:11 brouard
354: *** empty log message ***
355:
1.147 brouard 356: Revision 1.146 2014/06/16 10:20:28 brouard
357: Summary: Merge
358: Author: Brouard
359:
360: Merge, before building revised version.
361:
1.146 brouard 362: Revision 1.145 2014/06/10 21:23:15 brouard
363: Summary: Debugging with valgrind
364: Author: Nicolas Brouard
365:
366: Lot of changes in order to output the results with some covariates
367: After the Edimburgh REVES conference 2014, it seems mandatory to
368: improve the code.
369: No more memory valgrind error but a lot has to be done in order to
370: continue the work of splitting the code into subroutines.
371: Also, decodemodel has been improved. Tricode is still not
372: optimal. nbcode should be improved. Documentation has been added in
373: the source code.
374:
1.144 brouard 375: Revision 1.143 2014/01/26 09:45:38 brouard
376: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
377:
378: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
379: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
380:
1.143 brouard 381: Revision 1.142 2014/01/26 03:57:36 brouard
382: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
383:
384: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
385:
1.142 brouard 386: Revision 1.141 2014/01/26 02:42:01 brouard
387: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
388:
1.141 brouard 389: Revision 1.140 2011/09/02 10:37:54 brouard
390: Summary: times.h is ok with mingw32 now.
391:
1.140 brouard 392: Revision 1.139 2010/06/14 07:50:17 brouard
393: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
394: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
395:
1.139 brouard 396: Revision 1.138 2010/04/30 18:19:40 brouard
397: *** empty log message ***
398:
1.138 brouard 399: Revision 1.137 2010/04/29 18:11:38 brouard
400: (Module): Checking covariates for more complex models
401: than V1+V2. A lot of change to be done. Unstable.
402:
1.137 brouard 403: Revision 1.136 2010/04/26 20:30:53 brouard
404: (Module): merging some libgsl code. Fixing computation
405: of likelione (using inter/intrapolation if mle = 0) in order to
406: get same likelihood as if mle=1.
407: Some cleaning of code and comments added.
408:
1.136 brouard 409: Revision 1.135 2009/10/29 15:33:14 brouard
410: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
411:
1.135 brouard 412: Revision 1.134 2009/10/29 13:18:53 brouard
413: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
414:
1.134 brouard 415: Revision 1.133 2009/07/06 10:21:25 brouard
416: just nforces
417:
1.133 brouard 418: Revision 1.132 2009/07/06 08:22:05 brouard
419: Many tings
420:
1.132 brouard 421: Revision 1.131 2009/06/20 16:22:47 brouard
422: Some dimensions resccaled
423:
1.131 brouard 424: Revision 1.130 2009/05/26 06:44:34 brouard
425: (Module): Max Covariate is now set to 20 instead of 8. A
426: lot of cleaning with variables initialized to 0. Trying to make
427: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
428:
1.130 brouard 429: Revision 1.129 2007/08/31 13:49:27 lievre
430: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
431:
1.129 lievre 432: Revision 1.128 2006/06/30 13:02:05 brouard
433: (Module): Clarifications on computing e.j
434:
1.128 brouard 435: Revision 1.127 2006/04/28 18:11:50 brouard
436: (Module): Yes the sum of survivors was wrong since
437: imach-114 because nhstepm was no more computed in the age
438: loop. Now we define nhstepma in the age loop.
439: (Module): In order to speed up (in case of numerous covariates) we
440: compute health expectancies (without variances) in a first step
441: and then all the health expectancies with variances or standard
442: deviation (needs data from the Hessian matrices) which slows the
443: computation.
444: In the future we should be able to stop the program is only health
445: expectancies and graph are needed without standard deviations.
446:
1.127 brouard 447: Revision 1.126 2006/04/28 17:23:28 brouard
448: (Module): Yes the sum of survivors was wrong since
449: imach-114 because nhstepm was no more computed in the age
450: loop. Now we define nhstepma in the age loop.
451: Version 0.98h
452:
1.126 brouard 453: Revision 1.125 2006/04/04 15:20:31 lievre
454: Errors in calculation of health expectancies. Age was not initialized.
455: Forecasting file added.
456:
457: Revision 1.124 2006/03/22 17:13:53 lievre
458: Parameters are printed with %lf instead of %f (more numbers after the comma).
459: The log-likelihood is printed in the log file
460:
461: Revision 1.123 2006/03/20 10:52:43 brouard
462: * imach.c (Module): <title> changed, corresponds to .htm file
463: name. <head> headers where missing.
464:
465: * imach.c (Module): Weights can have a decimal point as for
466: English (a comma might work with a correct LC_NUMERIC environment,
467: otherwise the weight is truncated).
468: Modification of warning when the covariates values are not 0 or
469: 1.
470: Version 0.98g
471:
472: Revision 1.122 2006/03/20 09:45:41 brouard
473: (Module): Weights can have a decimal point as for
474: English (a comma might work with a correct LC_NUMERIC environment,
475: otherwise the weight is truncated).
476: Modification of warning when the covariates values are not 0 or
477: 1.
478: Version 0.98g
479:
480: Revision 1.121 2006/03/16 17:45:01 lievre
481: * imach.c (Module): Comments concerning covariates added
482:
483: * imach.c (Module): refinements in the computation of lli if
484: status=-2 in order to have more reliable computation if stepm is
485: not 1 month. Version 0.98f
486:
487: Revision 1.120 2006/03/16 15:10:38 lievre
488: (Module): refinements in the computation of lli if
489: status=-2 in order to have more reliable computation if stepm is
490: not 1 month. Version 0.98f
491:
492: Revision 1.119 2006/03/15 17:42:26 brouard
493: (Module): Bug if status = -2, the loglikelihood was
494: computed as likelihood omitting the logarithm. Version O.98e
495:
496: Revision 1.118 2006/03/14 18:20:07 brouard
497: (Module): varevsij Comments added explaining the second
498: table of variances if popbased=1 .
499: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
500: (Module): Function pstamp added
501: (Module): Version 0.98d
502:
503: Revision 1.117 2006/03/14 17:16:22 brouard
504: (Module): varevsij Comments added explaining the second
505: table of variances if popbased=1 .
506: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
507: (Module): Function pstamp added
508: (Module): Version 0.98d
509:
510: Revision 1.116 2006/03/06 10:29:27 brouard
511: (Module): Variance-covariance wrong links and
512: varian-covariance of ej. is needed (Saito).
513:
514: Revision 1.115 2006/02/27 12:17:45 brouard
515: (Module): One freematrix added in mlikeli! 0.98c
516:
517: Revision 1.114 2006/02/26 12:57:58 brouard
518: (Module): Some improvements in processing parameter
519: filename with strsep.
520:
521: Revision 1.113 2006/02/24 14:20:24 brouard
522: (Module): Memory leaks checks with valgrind and:
523: datafile was not closed, some imatrix were not freed and on matrix
524: allocation too.
525:
526: Revision 1.112 2006/01/30 09:55:26 brouard
527: (Module): Back to gnuplot.exe instead of wgnuplot.exe
528:
529: Revision 1.111 2006/01/25 20:38:18 brouard
530: (Module): Lots of cleaning and bugs added (Gompertz)
531: (Module): Comments can be added in data file. Missing date values
532: can be a simple dot '.'.
533:
534: Revision 1.110 2006/01/25 00:51:50 brouard
535: (Module): Lots of cleaning and bugs added (Gompertz)
536:
537: Revision 1.109 2006/01/24 19:37:15 brouard
538: (Module): Comments (lines starting with a #) are allowed in data.
539:
540: Revision 1.108 2006/01/19 18:05:42 lievre
541: Gnuplot problem appeared...
542: To be fixed
543:
544: Revision 1.107 2006/01/19 16:20:37 brouard
545: Test existence of gnuplot in imach path
546:
547: Revision 1.106 2006/01/19 13:24:36 brouard
548: Some cleaning and links added in html output
549:
550: Revision 1.105 2006/01/05 20:23:19 lievre
551: *** empty log message ***
552:
553: Revision 1.104 2005/09/30 16:11:43 lievre
554: (Module): sump fixed, loop imx fixed, and simplifications.
555: (Module): If the status is missing at the last wave but we know
556: that the person is alive, then we can code his/her status as -2
557: (instead of missing=-1 in earlier versions) and his/her
558: contributions to the likelihood is 1 - Prob of dying from last
559: health status (= 1-p13= p11+p12 in the easiest case of somebody in
560: the healthy state at last known wave). Version is 0.98
561:
562: Revision 1.103 2005/09/30 15:54:49 lievre
563: (Module): sump fixed, loop imx fixed, and simplifications.
564:
565: Revision 1.102 2004/09/15 17:31:30 brouard
566: Add the possibility to read data file including tab characters.
567:
568: Revision 1.101 2004/09/15 10:38:38 brouard
569: Fix on curr_time
570:
571: Revision 1.100 2004/07/12 18:29:06 brouard
572: Add version for Mac OS X. Just define UNIX in Makefile
573:
574: Revision 1.99 2004/06/05 08:57:40 brouard
575: *** empty log message ***
576:
577: Revision 1.98 2004/05/16 15:05:56 brouard
578: New version 0.97 . First attempt to estimate force of mortality
579: directly from the data i.e. without the need of knowing the health
580: state at each age, but using a Gompertz model: log u =a + b*age .
581: This is the basic analysis of mortality and should be done before any
582: other analysis, in order to test if the mortality estimated from the
583: cross-longitudinal survey is different from the mortality estimated
584: from other sources like vital statistic data.
585:
586: The same imach parameter file can be used but the option for mle should be -3.
587:
1.133 brouard 588: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 589: former routines in order to include the new code within the former code.
590:
591: The output is very simple: only an estimate of the intercept and of
592: the slope with 95% confident intervals.
593:
594: Current limitations:
595: A) Even if you enter covariates, i.e. with the
596: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
597: B) There is no computation of Life Expectancy nor Life Table.
598:
599: Revision 1.97 2004/02/20 13:25:42 lievre
600: Version 0.96d. Population forecasting command line is (temporarily)
601: suppressed.
602:
603: Revision 1.96 2003/07/15 15:38:55 brouard
604: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
605: rewritten within the same printf. Workaround: many printfs.
606:
607: Revision 1.95 2003/07/08 07:54:34 brouard
608: * imach.c (Repository):
609: (Repository): Using imachwizard code to output a more meaningful covariance
610: matrix (cov(a12,c31) instead of numbers.
611:
612: Revision 1.94 2003/06/27 13:00:02 brouard
613: Just cleaning
614:
615: Revision 1.93 2003/06/25 16:33:55 brouard
616: (Module): On windows (cygwin) function asctime_r doesn't
617: exist so I changed back to asctime which exists.
618: (Module): Version 0.96b
619:
620: Revision 1.92 2003/06/25 16:30:45 brouard
621: (Module): On windows (cygwin) function asctime_r doesn't
622: exist so I changed back to asctime which exists.
623:
624: Revision 1.91 2003/06/25 15:30:29 brouard
625: * imach.c (Repository): Duplicated warning errors corrected.
626: (Repository): Elapsed time after each iteration is now output. It
627: helps to forecast when convergence will be reached. Elapsed time
628: is stamped in powell. We created a new html file for the graphs
629: concerning matrix of covariance. It has extension -cov.htm.
630:
631: Revision 1.90 2003/06/24 12:34:15 brouard
632: (Module): Some bugs corrected for windows. Also, when
633: mle=-1 a template is output in file "or"mypar.txt with the design
634: of the covariance matrix to be input.
635:
636: Revision 1.89 2003/06/24 12:30:52 brouard
637: (Module): Some bugs corrected for windows. Also, when
638: mle=-1 a template is output in file "or"mypar.txt with the design
639: of the covariance matrix to be input.
640:
641: Revision 1.88 2003/06/23 17:54:56 brouard
642: * 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.
643:
644: Revision 1.87 2003/06/18 12:26:01 brouard
645: Version 0.96
646:
647: Revision 1.86 2003/06/17 20:04:08 brouard
648: (Module): Change position of html and gnuplot routines and added
649: routine fileappend.
650:
651: Revision 1.85 2003/06/17 13:12:43 brouard
652: * imach.c (Repository): Check when date of death was earlier that
653: current date of interview. It may happen when the death was just
654: prior to the death. In this case, dh was negative and likelihood
655: was wrong (infinity). We still send an "Error" but patch by
656: assuming that the date of death was just one stepm after the
657: interview.
658: (Repository): Because some people have very long ID (first column)
659: we changed int to long in num[] and we added a new lvector for
660: memory allocation. But we also truncated to 8 characters (left
661: truncation)
662: (Repository): No more line truncation errors.
663:
664: Revision 1.84 2003/06/13 21:44:43 brouard
665: * imach.c (Repository): Replace "freqsummary" at a correct
666: place. It differs from routine "prevalence" which may be called
667: many times. Probs is memory consuming and must be used with
668: parcimony.
669: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
670:
671: Revision 1.83 2003/06/10 13:39:11 lievre
672: *** empty log message ***
673:
674: Revision 1.82 2003/06/05 15:57:20 brouard
675: Add log in imach.c and fullversion number is now printed.
676:
677: */
678: /*
679: Interpolated Markov Chain
680:
681: Short summary of the programme:
682:
1.227 brouard 683: This program computes Healthy Life Expectancies or State-specific
684: (if states aren't health statuses) Expectancies from
685: cross-longitudinal data. Cross-longitudinal data consist in:
686:
687: -1- a first survey ("cross") where individuals from different ages
688: are interviewed on their health status or degree of disability (in
689: the case of a health survey which is our main interest)
690:
691: -2- at least a second wave of interviews ("longitudinal") which
692: measure each change (if any) in individual health status. Health
693: expectancies are computed from the time spent in each health state
694: according to a model. More health states you consider, more time is
695: necessary to reach the Maximum Likelihood of the parameters involved
696: in the model. The simplest model is the multinomial logistic model
697: where pij is the probability to be observed in state j at the second
698: wave conditional to be observed in state i at the first
699: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
700: etc , where 'age' is age and 'sex' is a covariate. If you want to
701: have a more complex model than "constant and age", you should modify
702: the program where the markup *Covariates have to be included here
703: again* invites you to do it. More covariates you add, slower the
1.126 brouard 704: convergence.
705:
706: The advantage of this computer programme, compared to a simple
707: multinomial logistic model, is clear when the delay between waves is not
708: identical for each individual. Also, if a individual missed an
709: intermediate interview, the information is lost, but taken into
710: account using an interpolation or extrapolation.
711:
712: hPijx is the probability to be observed in state i at age x+h
713: conditional to the observed state i at age x. The delay 'h' can be
714: split into an exact number (nh*stepm) of unobserved intermediate
715: states. This elementary transition (by month, quarter,
716: semester or year) is modelled as a multinomial logistic. The hPx
717: matrix is simply the matrix product of nh*stepm elementary matrices
718: and the contribution of each individual to the likelihood is simply
719: hPijx.
720:
721: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 722: of the life expectancies. It also computes the period (stable) prevalence.
723:
724: Back prevalence and projections:
1.227 brouard 725:
726: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
727: double agemaxpar, double ftolpl, int *ncvyearp, double
728: dateprev1,double dateprev2, int firstpass, int lastpass, int
729: mobilavproj)
730:
731: Computes the back prevalence limit for any combination of
732: covariate values k at any age between ageminpar and agemaxpar and
733: returns it in **bprlim. In the loops,
734:
735: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
736: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
737:
738: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 739: Computes for any combination of covariates k and any age between bage and fage
740: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
741: oldm=oldms;savm=savms;
1.227 brouard 742:
743: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 744: Computes the transition matrix starting at age 'age' over
745: 'nhstepm*hstepm*stepm' months (i.e. until
746: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 747: nhstepm*hstepm matrices.
748:
749: Returns p3mat[i][j][h] after calling
750: p3mat[i][j][h]=matprod2(newm,
751: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
752: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
753: oldm);
1.226 brouard 754:
755: Important routines
756:
757: - func (or funcone), computes logit (pij) distinguishing
758: o fixed variables (single or product dummies or quantitative);
759: o varying variables by:
760: (1) wave (single, product dummies, quantitative),
761: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
762: % fixed dummy (treated) or quantitative (not done because time-consuming);
763: % varying dummy (not done) or quantitative (not done);
764: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
765: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
766: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
767: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
768: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 769:
1.226 brouard 770:
771:
1.133 brouard 772: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
773: Institut national d'études démographiques, Paris.
1.126 brouard 774: This software have been partly granted by Euro-REVES, a concerted action
775: from the European Union.
776: It is copyrighted identically to a GNU software product, ie programme and
777: software can be distributed freely for non commercial use. Latest version
778: can be accessed at http://euroreves.ined.fr/imach .
779:
780: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
781: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
782:
783: **********************************************************************/
784: /*
785: main
786: read parameterfile
787: read datafile
788: concatwav
789: freqsummary
790: if (mle >= 1)
791: mlikeli
792: print results files
793: if mle==1
794: computes hessian
795: read end of parameter file: agemin, agemax, bage, fage, estepm
796: begin-prev-date,...
797: open gnuplot file
798: open html file
1.145 brouard 799: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
800: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
801: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
802: freexexit2 possible for memory heap.
803:
804: h Pij x | pij_nom ficrestpij
805: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
806: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
807: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
808:
809: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
810: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
811: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
812: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
813: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
814:
1.126 brouard 815: forecasting if prevfcast==1 prevforecast call prevalence()
816: health expectancies
817: Variance-covariance of DFLE
818: prevalence()
819: movingaverage()
820: varevsij()
821: if popbased==1 varevsij(,popbased)
822: total life expectancies
823: Variance of period (stable) prevalence
824: end
825: */
826:
1.187 brouard 827: /* #define DEBUG */
828: /* #define DEBUGBRENT */
1.203 brouard 829: /* #define DEBUGLINMIN */
830: /* #define DEBUGHESS */
831: #define DEBUGHESSIJ
1.224 brouard 832: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 833: #define POWELL /* Instead of NLOPT */
1.224 brouard 834: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 835: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
836: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 837:
838: #include <math.h>
839: #include <stdio.h>
840: #include <stdlib.h>
841: #include <string.h>
1.226 brouard 842: #include <ctype.h>
1.159 brouard 843:
844: #ifdef _WIN32
845: #include <io.h>
1.172 brouard 846: #include <windows.h>
847: #include <tchar.h>
1.159 brouard 848: #else
1.126 brouard 849: #include <unistd.h>
1.159 brouard 850: #endif
1.126 brouard 851:
852: #include <limits.h>
853: #include <sys/types.h>
1.171 brouard 854:
855: #if defined(__GNUC__)
856: #include <sys/utsname.h> /* Doesn't work on Windows */
857: #endif
858:
1.126 brouard 859: #include <sys/stat.h>
860: #include <errno.h>
1.159 brouard 861: /* extern int errno; */
1.126 brouard 862:
1.157 brouard 863: /* #ifdef LINUX */
864: /* #include <time.h> */
865: /* #include "timeval.h" */
866: /* #else */
867: /* #include <sys/time.h> */
868: /* #endif */
869:
1.126 brouard 870: #include <time.h>
871:
1.136 brouard 872: #ifdef GSL
873: #include <gsl/gsl_errno.h>
874: #include <gsl/gsl_multimin.h>
875: #endif
876:
1.167 brouard 877:
1.162 brouard 878: #ifdef NLOPT
879: #include <nlopt.h>
880: typedef struct {
881: double (* function)(double [] );
882: } myfunc_data ;
883: #endif
884:
1.126 brouard 885: /* #include <libintl.h> */
886: /* #define _(String) gettext (String) */
887:
1.141 brouard 888: #define MAXLINE 1024 /* Was 256. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 889:
890: #define GNUPLOTPROGRAM "gnuplot"
891: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
892: #define FILENAMELENGTH 132
893:
894: #define GLOCK_ERROR_NOPATH -1 /* empty path */
895: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
896:
1.144 brouard 897: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
898: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 899:
900: #define NINTERVMAX 8
1.144 brouard 901: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
902: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
903: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 904: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 905: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
906: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 907: #define MAXN 20000
1.144 brouard 908: #define YEARM 12. /**< Number of months per year */
1.218 brouard 909: /* #define AGESUP 130 */
910: #define AGESUP 150
911: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 912: #define AGEBASE 40
1.194 brouard 913: #define AGEOVERFLOW 1.e20
1.164 brouard 914: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 915: #ifdef _WIN32
916: #define DIRSEPARATOR '\\'
917: #define CHARSEPARATOR "\\"
918: #define ODIRSEPARATOR '/'
919: #else
1.126 brouard 920: #define DIRSEPARATOR '/'
921: #define CHARSEPARATOR "/"
922: #define ODIRSEPARATOR '\\'
923: #endif
924:
1.241 ! brouard 925: /* $Id: imach.c,v 1.240 2016/08/29 07:53:18 brouard Exp $ */
1.126 brouard 926: /* $State: Exp $ */
1.196 brouard 927: #include "version.h"
928: char version[]=__IMACH_VERSION__;
1.224 brouard 929: 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";
1.241 ! brouard 930: char fullversion[]="$Revision: 1.240 $ $Date: 2016/08/29 07:53:18 $";
1.126 brouard 931: char strstart[80];
932: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 933: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 934: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 935: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
936: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
937: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 938: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
939: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 940: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
941: int cptcovprodnoage=0; /**< Number of covariate products without age */
942: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 943: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
944: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 945: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 946: int nsd=0; /**< Total number of single dummy variables (output) */
947: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 948: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 949: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 950: int ntveff=0; /**< ntveff number of effective time varying variables */
951: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 952: int cptcov=0; /* Working variable */
1.218 brouard 953: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 954: int npar=NPARMAX;
955: int nlstate=2; /* Number of live states */
956: int ndeath=1; /* Number of dead states */
1.130 brouard 957: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 958: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 959: int popbased=0;
960:
961: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 962: int maxwav=0; /* Maxim number of waves */
963: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
964: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
965: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 966: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 967: int mle=1, weightopt=0;
1.126 brouard 968: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
969: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
970: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
971: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 972: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 973: int selected(int kvar); /* Is covariate kvar selected for printing results */
974:
1.130 brouard 975: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 976: double **matprod2(); /* test */
1.126 brouard 977: double **oldm, **newm, **savm; /* Working pointers to matrices */
978: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 979: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
980:
1.136 brouard 981: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 982: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 983: FILE *ficlog, *ficrespow;
1.130 brouard 984: int globpr=0; /* Global variable for printing or not */
1.126 brouard 985: double fretone; /* Only one call to likelihood */
1.130 brouard 986: long ipmx=0; /* Number of contributions */
1.126 brouard 987: double sw; /* Sum of weights */
988: char filerespow[FILENAMELENGTH];
989: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
990: FILE *ficresilk;
991: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
992: FILE *ficresprobmorprev;
993: FILE *fichtm, *fichtmcov; /* Html File */
994: FILE *ficreseij;
995: char filerese[FILENAMELENGTH];
996: FILE *ficresstdeij;
997: char fileresstde[FILENAMELENGTH];
998: FILE *ficrescveij;
999: char filerescve[FILENAMELENGTH];
1000: FILE *ficresvij;
1001: char fileresv[FILENAMELENGTH];
1002: FILE *ficresvpl;
1003: char fileresvpl[FILENAMELENGTH];
1004: char title[MAXLINE];
1.234 brouard 1005: char model[MAXLINE]; /**< The model line */
1.217 brouard 1006: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1007: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1008: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1009: char command[FILENAMELENGTH];
1010: int outcmd=0;
1011:
1.217 brouard 1012: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1013: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1014: char filelog[FILENAMELENGTH]; /* Log file */
1015: char filerest[FILENAMELENGTH];
1016: char fileregp[FILENAMELENGTH];
1017: char popfile[FILENAMELENGTH];
1018:
1019: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1020:
1.157 brouard 1021: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1022: /* struct timezone tzp; */
1023: /* extern int gettimeofday(); */
1024: struct tm tml, *gmtime(), *localtime();
1025:
1026: extern time_t time();
1027:
1028: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1029: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1030: struct tm tm;
1031:
1.126 brouard 1032: char strcurr[80], strfor[80];
1033:
1034: char *endptr;
1035: long lval;
1036: double dval;
1037:
1038: #define NR_END 1
1039: #define FREE_ARG char*
1040: #define FTOL 1.0e-10
1041:
1042: #define NRANSI
1.240 brouard 1043: #define ITMAX 200
1044: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1045:
1046: #define TOL 2.0e-4
1047:
1048: #define CGOLD 0.3819660
1049: #define ZEPS 1.0e-10
1050: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1051:
1052: #define GOLD 1.618034
1053: #define GLIMIT 100.0
1054: #define TINY 1.0e-20
1055:
1056: static double maxarg1,maxarg2;
1057: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1058: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1059:
1060: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1061: #define rint(a) floor(a+0.5)
1.166 brouard 1062: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1063: #define mytinydouble 1.0e-16
1.166 brouard 1064: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1065: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1066: /* static double dsqrarg; */
1067: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1068: static double sqrarg;
1069: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1070: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1071: int agegomp= AGEGOMP;
1072:
1073: int imx;
1074: int stepm=1;
1075: /* Stepm, step in month: minimum step interpolation*/
1076:
1077: int estepm;
1078: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1079:
1080: int m,nb;
1081: long *num;
1.197 brouard 1082: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1083: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1084: covariate for which somebody answered excluding
1085: undefined. Usually 2: 0 and 1. */
1086: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1087: covariate for which somebody answered including
1088: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1089: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1090: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1091: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1092: double *ageexmed,*agecens;
1093: double dateintmean=0;
1094:
1095: double *weight;
1096: int **s; /* Status */
1.141 brouard 1097: double *agedc;
1.145 brouard 1098: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1099: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1100: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1101: double **coqvar; /* Fixed quantitative covariate iqv */
1102: double ***cotvar; /* Time varying covariate itv */
1103: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1104: double idx;
1105: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1106: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1107: /*k 1 2 3 4 5 6 7 8 9 */
1108: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1109: /* Tndvar[k] 1 2 3 4 5 */
1110: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1111: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1112: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1113: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1114: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1115: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1116: /* Tprod[i]=k 4 7 */
1117: /* Tage[i]=k 5 8 */
1118: /* */
1119: /* Type */
1120: /* V 1 2 3 4 5 */
1121: /* F F V V V */
1122: /* D Q D D Q */
1123: /* */
1124: int *TvarsD;
1125: int *TvarsDind;
1126: int *TvarsQ;
1127: int *TvarsQind;
1128:
1.235 brouard 1129: #define MAXRESULTLINES 10
1130: int nresult=0;
1131: int TKresult[MAXRESULTLINES];
1.237 brouard 1132: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1133: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1134: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1135: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1136: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1137: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1138:
1.234 brouard 1139: /* 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 1140: 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 */
1141: 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 */
1142: 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 */
1143: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1144: 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 */
1145: 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 1146: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1147: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1148: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1149: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1150: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1151: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1152: 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 */
1153: 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 */
1154:
1.230 brouard 1155: int *Tvarsel; /**< Selected covariates for output */
1156: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1157: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1158: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1159: 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 1160: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1161: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1162: int *Tage;
1.227 brouard 1163: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1164: 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 1165: 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*/
1166: 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 1167: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1168: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1169: int **Tvard;
1170: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1171: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1172: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1173: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1174: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1175: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1176: double *lsurv, *lpop, *tpop;
1177:
1.231 brouard 1178: #define FD 1; /* Fixed dummy covariate */
1179: #define FQ 2; /* Fixed quantitative covariate */
1180: #define FP 3; /* Fixed product covariate */
1181: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1182: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1183: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1184: #define VD 10; /* Varying dummy covariate */
1185: #define VQ 11; /* Varying quantitative covariate */
1186: #define VP 12; /* Varying product covariate */
1187: #define VPDD 13; /* Varying product dummy*dummy covariate */
1188: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1189: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1190: #define APFD 16; /* Age product * fixed dummy covariate */
1191: #define APFQ 17; /* Age product * fixed quantitative covariate */
1192: #define APVD 18; /* Age product * varying dummy covariate */
1193: #define APVQ 19; /* Age product * varying quantitative covariate */
1194:
1195: #define FTYPE 1; /* Fixed covariate */
1196: #define VTYPE 2; /* Varying covariate (loop in wave) */
1197: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1198:
1199: struct kmodel{
1200: int maintype; /* main type */
1201: int subtype; /* subtype */
1202: };
1203: struct kmodel modell[NCOVMAX];
1204:
1.143 brouard 1205: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1206: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1207:
1208: /**************** split *************************/
1209: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1210: {
1211: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1212: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1213: */
1214: char *ss; /* pointer */
1.186 brouard 1215: int l1=0, l2=0; /* length counters */
1.126 brouard 1216:
1217: l1 = strlen(path ); /* length of path */
1218: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1219: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1220: if ( ss == NULL ) { /* no directory, so determine current directory */
1221: strcpy( name, path ); /* we got the fullname name because no directory */
1222: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1223: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1224: /* get current working directory */
1225: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1226: #ifdef WIN32
1227: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1228: #else
1229: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1230: #endif
1.126 brouard 1231: return( GLOCK_ERROR_GETCWD );
1232: }
1233: /* got dirc from getcwd*/
1234: printf(" DIRC = %s \n",dirc);
1.205 brouard 1235: } else { /* strip directory from path */
1.126 brouard 1236: ss++; /* after this, the filename */
1237: l2 = strlen( ss ); /* length of filename */
1238: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1239: strcpy( name, ss ); /* save file name */
1240: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1241: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1242: printf(" DIRC2 = %s \n",dirc);
1243: }
1244: /* We add a separator at the end of dirc if not exists */
1245: l1 = strlen( dirc ); /* length of directory */
1246: if( dirc[l1-1] != DIRSEPARATOR ){
1247: dirc[l1] = DIRSEPARATOR;
1248: dirc[l1+1] = 0;
1249: printf(" DIRC3 = %s \n",dirc);
1250: }
1251: ss = strrchr( name, '.' ); /* find last / */
1252: if (ss >0){
1253: ss++;
1254: strcpy(ext,ss); /* save extension */
1255: l1= strlen( name);
1256: l2= strlen(ss)+1;
1257: strncpy( finame, name, l1-l2);
1258: finame[l1-l2]= 0;
1259: }
1260:
1261: return( 0 ); /* we're done */
1262: }
1263:
1264:
1265: /******************************************/
1266:
1267: void replace_back_to_slash(char *s, char*t)
1268: {
1269: int i;
1270: int lg=0;
1271: i=0;
1272: lg=strlen(t);
1273: for(i=0; i<= lg; i++) {
1274: (s[i] = t[i]);
1275: if (t[i]== '\\') s[i]='/';
1276: }
1277: }
1278:
1.132 brouard 1279: char *trimbb(char *out, char *in)
1.137 brouard 1280: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1281: char *s;
1282: s=out;
1283: while (*in != '\0'){
1.137 brouard 1284: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1285: in++;
1286: }
1287: *out++ = *in++;
1288: }
1289: *out='\0';
1290: return s;
1291: }
1292:
1.187 brouard 1293: /* char *substrchaine(char *out, char *in, char *chain) */
1294: /* { */
1295: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1296: /* char *s, *t; */
1297: /* t=in;s=out; */
1298: /* while ((*in != *chain) && (*in != '\0')){ */
1299: /* *out++ = *in++; */
1300: /* } */
1301:
1302: /* /\* *in matches *chain *\/ */
1303: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1304: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1305: /* } */
1306: /* in--; chain--; */
1307: /* while ( (*in != '\0')){ */
1308: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1309: /* *out++ = *in++; */
1310: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1311: /* } */
1312: /* *out='\0'; */
1313: /* out=s; */
1314: /* return out; */
1315: /* } */
1316: char *substrchaine(char *out, char *in, char *chain)
1317: {
1318: /* Substract chain 'chain' from 'in', return and output 'out' */
1319: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1320:
1321: char *strloc;
1322:
1323: strcpy (out, in);
1324: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1325: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1326: if(strloc != NULL){
1327: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1328: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1329: /* strcpy (strloc, strloc +strlen(chain));*/
1330: }
1331: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1332: return out;
1333: }
1334:
1335:
1.145 brouard 1336: char *cutl(char *blocc, char *alocc, char *in, char occ)
1337: {
1.187 brouard 1338: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1339: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1340: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1341: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1342: */
1.160 brouard 1343: char *s, *t;
1.145 brouard 1344: t=in;s=in;
1345: while ((*in != occ) && (*in != '\0')){
1346: *alocc++ = *in++;
1347: }
1348: if( *in == occ){
1349: *(alocc)='\0';
1350: s=++in;
1351: }
1352:
1353: if (s == t) {/* occ not found */
1354: *(alocc-(in-s))='\0';
1355: in=s;
1356: }
1357: while ( *in != '\0'){
1358: *blocc++ = *in++;
1359: }
1360:
1361: *blocc='\0';
1362: return t;
1363: }
1.137 brouard 1364: char *cutv(char *blocc, char *alocc, char *in, char occ)
1365: {
1.187 brouard 1366: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1367: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1368: gives blocc="abcdef2ghi" and alocc="j".
1369: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1370: */
1371: char *s, *t;
1372: t=in;s=in;
1373: while (*in != '\0'){
1374: while( *in == occ){
1375: *blocc++ = *in++;
1376: s=in;
1377: }
1378: *blocc++ = *in++;
1379: }
1380: if (s == t) /* occ not found */
1381: *(blocc-(in-s))='\0';
1382: else
1383: *(blocc-(in-s)-1)='\0';
1384: in=s;
1385: while ( *in != '\0'){
1386: *alocc++ = *in++;
1387: }
1388:
1389: *alocc='\0';
1390: return s;
1391: }
1392:
1.126 brouard 1393: int nbocc(char *s, char occ)
1394: {
1395: int i,j=0;
1396: int lg=20;
1397: i=0;
1398: lg=strlen(s);
1399: for(i=0; i<= lg; i++) {
1.234 brouard 1400: if (s[i] == occ ) j++;
1.126 brouard 1401: }
1402: return j;
1403: }
1404:
1.137 brouard 1405: /* void cutv(char *u,char *v, char*t, char occ) */
1406: /* { */
1407: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1408: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1409: /* gives u="abcdef2ghi" and v="j" *\/ */
1410: /* int i,lg,j,p=0; */
1411: /* i=0; */
1412: /* lg=strlen(t); */
1413: /* for(j=0; j<=lg-1; j++) { */
1414: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1415: /* } */
1.126 brouard 1416:
1.137 brouard 1417: /* for(j=0; j<p; j++) { */
1418: /* (u[j] = t[j]); */
1419: /* } */
1420: /* u[p]='\0'; */
1.126 brouard 1421:
1.137 brouard 1422: /* for(j=0; j<= lg; j++) { */
1423: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1424: /* } */
1425: /* } */
1.126 brouard 1426:
1.160 brouard 1427: #ifdef _WIN32
1428: char * strsep(char **pp, const char *delim)
1429: {
1430: char *p, *q;
1431:
1432: if ((p = *pp) == NULL)
1433: return 0;
1434: if ((q = strpbrk (p, delim)) != NULL)
1435: {
1436: *pp = q + 1;
1437: *q = '\0';
1438: }
1439: else
1440: *pp = 0;
1441: return p;
1442: }
1443: #endif
1444:
1.126 brouard 1445: /********************** nrerror ********************/
1446:
1447: void nrerror(char error_text[])
1448: {
1449: fprintf(stderr,"ERREUR ...\n");
1450: fprintf(stderr,"%s\n",error_text);
1451: exit(EXIT_FAILURE);
1452: }
1453: /*********************** vector *******************/
1454: double *vector(int nl, int nh)
1455: {
1456: double *v;
1457: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1458: if (!v) nrerror("allocation failure in vector");
1459: return v-nl+NR_END;
1460: }
1461:
1462: /************************ free vector ******************/
1463: void free_vector(double*v, int nl, int nh)
1464: {
1465: free((FREE_ARG)(v+nl-NR_END));
1466: }
1467:
1468: /************************ivector *******************************/
1469: int *ivector(long nl,long nh)
1470: {
1471: int *v;
1472: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1473: if (!v) nrerror("allocation failure in ivector");
1474: return v-nl+NR_END;
1475: }
1476:
1477: /******************free ivector **************************/
1478: void free_ivector(int *v, long nl, long nh)
1479: {
1480: free((FREE_ARG)(v+nl-NR_END));
1481: }
1482:
1483: /************************lvector *******************************/
1484: long *lvector(long nl,long nh)
1485: {
1486: long *v;
1487: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1488: if (!v) nrerror("allocation failure in ivector");
1489: return v-nl+NR_END;
1490: }
1491:
1492: /******************free lvector **************************/
1493: void free_lvector(long *v, long nl, long nh)
1494: {
1495: free((FREE_ARG)(v+nl-NR_END));
1496: }
1497:
1498: /******************* imatrix *******************************/
1499: int **imatrix(long nrl, long nrh, long ncl, long nch)
1500: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1501: {
1502: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1503: int **m;
1504:
1505: /* allocate pointers to rows */
1506: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1507: if (!m) nrerror("allocation failure 1 in matrix()");
1508: m += NR_END;
1509: m -= nrl;
1510:
1511:
1512: /* allocate rows and set pointers to them */
1513: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1514: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1515: m[nrl] += NR_END;
1516: m[nrl] -= ncl;
1517:
1518: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1519:
1520: /* return pointer to array of pointers to rows */
1521: return m;
1522: }
1523:
1524: /****************** free_imatrix *************************/
1525: void free_imatrix(m,nrl,nrh,ncl,nch)
1526: int **m;
1527: long nch,ncl,nrh,nrl;
1528: /* free an int matrix allocated by imatrix() */
1529: {
1530: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1531: free((FREE_ARG) (m+nrl-NR_END));
1532: }
1533:
1534: /******************* matrix *******************************/
1535: double **matrix(long nrl, long nrh, long ncl, long nch)
1536: {
1537: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1538: double **m;
1539:
1540: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1541: if (!m) nrerror("allocation failure 1 in matrix()");
1542: m += NR_END;
1543: m -= nrl;
1544:
1545: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1546: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1547: m[nrl] += NR_END;
1548: m[nrl] -= ncl;
1549:
1550: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1551: return m;
1.145 brouard 1552: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1553: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1554: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1555: */
1556: }
1557:
1558: /*************************free matrix ************************/
1559: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1560: {
1561: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1562: free((FREE_ARG)(m+nrl-NR_END));
1563: }
1564:
1565: /******************* ma3x *******************************/
1566: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1567: {
1568: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+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:
1583: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1584: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1585: m[nrl][ncl] += NR_END;
1586: m[nrl][ncl] -= nll;
1587: for (j=ncl+1; j<=nch; j++)
1588: m[nrl][j]=m[nrl][j-1]+nlay;
1589:
1590: for (i=nrl+1; i<=nrh; i++) {
1591: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1592: for (j=ncl+1; j<=nch; j++)
1593: m[i][j]=m[i][j-1]+nlay;
1594: }
1595: return m;
1596: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1597: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1598: */
1599: }
1600:
1601: /*************************free ma3x ************************/
1602: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1603: {
1604: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1605: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1606: free((FREE_ARG)(m+nrl-NR_END));
1607: }
1608:
1609: /*************** function subdirf ***********/
1610: char *subdirf(char fileres[])
1611: {
1612: /* Caution optionfilefiname is hidden */
1613: strcpy(tmpout,optionfilefiname);
1614: strcat(tmpout,"/"); /* Add to the right */
1615: strcat(tmpout,fileres);
1616: return tmpout;
1617: }
1618:
1619: /*************** function subdirf2 ***********/
1620: char *subdirf2(char fileres[], char *preop)
1621: {
1622:
1623: /* Caution optionfilefiname is hidden */
1624: strcpy(tmpout,optionfilefiname);
1625: strcat(tmpout,"/");
1626: strcat(tmpout,preop);
1627: strcat(tmpout,fileres);
1628: return tmpout;
1629: }
1630:
1631: /*************** function subdirf3 ***********/
1632: char *subdirf3(char fileres[], char *preop, char *preop2)
1633: {
1634:
1635: /* Caution optionfilefiname is hidden */
1636: strcpy(tmpout,optionfilefiname);
1637: strcat(tmpout,"/");
1638: strcat(tmpout,preop);
1639: strcat(tmpout,preop2);
1640: strcat(tmpout,fileres);
1641: return tmpout;
1642: }
1.213 brouard 1643:
1644: /*************** function subdirfext ***********/
1645: char *subdirfext(char fileres[], char *preop, char *postop)
1646: {
1647:
1648: strcpy(tmpout,preop);
1649: strcat(tmpout,fileres);
1650: strcat(tmpout,postop);
1651: return tmpout;
1652: }
1.126 brouard 1653:
1.213 brouard 1654: /*************** function subdirfext3 ***********/
1655: char *subdirfext3(char fileres[], char *preop, char *postop)
1656: {
1657:
1658: /* Caution optionfilefiname is hidden */
1659: strcpy(tmpout,optionfilefiname);
1660: strcat(tmpout,"/");
1661: strcat(tmpout,preop);
1662: strcat(tmpout,fileres);
1663: strcat(tmpout,postop);
1664: return tmpout;
1665: }
1666:
1.162 brouard 1667: char *asc_diff_time(long time_sec, char ascdiff[])
1668: {
1669: long sec_left, days, hours, minutes;
1670: days = (time_sec) / (60*60*24);
1671: sec_left = (time_sec) % (60*60*24);
1672: hours = (sec_left) / (60*60) ;
1673: sec_left = (sec_left) %(60*60);
1674: minutes = (sec_left) /60;
1675: sec_left = (sec_left) % (60);
1676: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1677: return ascdiff;
1678: }
1679:
1.126 brouard 1680: /***************** f1dim *************************/
1681: extern int ncom;
1682: extern double *pcom,*xicom;
1683: extern double (*nrfunc)(double []);
1684:
1685: double f1dim(double x)
1686: {
1687: int j;
1688: double f;
1689: double *xt;
1690:
1691: xt=vector(1,ncom);
1692: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1693: f=(*nrfunc)(xt);
1694: free_vector(xt,1,ncom);
1695: return f;
1696: }
1697:
1698: /*****************brent *************************/
1699: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1700: {
1701: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1702: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1703: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1704: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1705: * returned function value.
1706: */
1.126 brouard 1707: int iter;
1708: double a,b,d,etemp;
1.159 brouard 1709: double fu=0,fv,fw,fx;
1.164 brouard 1710: double ftemp=0.;
1.126 brouard 1711: double p,q,r,tol1,tol2,u,v,w,x,xm;
1712: double e=0.0;
1713:
1714: a=(ax < cx ? ax : cx);
1715: b=(ax > cx ? ax : cx);
1716: x=w=v=bx;
1717: fw=fv=fx=(*f)(x);
1718: for (iter=1;iter<=ITMAX;iter++) {
1719: xm=0.5*(a+b);
1720: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1721: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1722: printf(".");fflush(stdout);
1723: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1724: #ifdef DEBUGBRENT
1.126 brouard 1725: 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);
1726: 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);
1727: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1728: #endif
1729: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1730: *xmin=x;
1731: return fx;
1732: }
1733: ftemp=fu;
1734: if (fabs(e) > tol1) {
1735: r=(x-w)*(fx-fv);
1736: q=(x-v)*(fx-fw);
1737: p=(x-v)*q-(x-w)*r;
1738: q=2.0*(q-r);
1739: if (q > 0.0) p = -p;
1740: q=fabs(q);
1741: etemp=e;
1742: e=d;
1743: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1744: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1745: else {
1.224 brouard 1746: d=p/q;
1747: u=x+d;
1748: if (u-a < tol2 || b-u < tol2)
1749: d=SIGN(tol1,xm-x);
1.126 brouard 1750: }
1751: } else {
1752: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1753: }
1754: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1755: fu=(*f)(u);
1756: if (fu <= fx) {
1757: if (u >= x) a=x; else b=x;
1758: SHFT(v,w,x,u)
1.183 brouard 1759: SHFT(fv,fw,fx,fu)
1760: } else {
1761: if (u < x) a=u; else b=u;
1762: if (fu <= fw || w == x) {
1.224 brouard 1763: v=w;
1764: w=u;
1765: fv=fw;
1766: fw=fu;
1.183 brouard 1767: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1768: v=u;
1769: fv=fu;
1.183 brouard 1770: }
1771: }
1.126 brouard 1772: }
1773: nrerror("Too many iterations in brent");
1774: *xmin=x;
1775: return fx;
1776: }
1777:
1778: /****************** mnbrak ***********************/
1779:
1780: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1781: double (*func)(double))
1.183 brouard 1782: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1783: the downhill direction (defined by the function as evaluated at the initial points) and returns
1784: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1785: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1786: */
1.126 brouard 1787: double ulim,u,r,q, dum;
1788: double fu;
1.187 brouard 1789:
1790: double scale=10.;
1791: int iterscale=0;
1792:
1793: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1794: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1795:
1796:
1797: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1798: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1799: /* *bx = *ax - (*ax - *bx)/scale; */
1800: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1801: /* } */
1802:
1.126 brouard 1803: if (*fb > *fa) {
1804: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1805: SHFT(dum,*fb,*fa,dum)
1806: }
1.126 brouard 1807: *cx=(*bx)+GOLD*(*bx-*ax);
1808: *fc=(*func)(*cx);
1.183 brouard 1809: #ifdef DEBUG
1.224 brouard 1810: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1811: 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 1812: #endif
1.224 brouard 1813: 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 1814: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1815: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1816: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1817: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1818: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1819: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1820: fu=(*func)(u);
1.163 brouard 1821: #ifdef DEBUG
1822: /* f(x)=A(x-u)**2+f(u) */
1823: double A, fparabu;
1824: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1825: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1826: 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);
1827: 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 1828: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1829: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1830: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1831: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1832: #endif
1.184 brouard 1833: #ifdef MNBRAKORIGINAL
1.183 brouard 1834: #else
1.191 brouard 1835: /* if (fu > *fc) { */
1836: /* #ifdef DEBUG */
1837: /* printf("mnbrak4 fu > fc \n"); */
1838: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1839: /* #endif */
1840: /* /\* 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 *\\/ *\/ */
1841: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1842: /* dum=u; /\* Shifting c and u *\/ */
1843: /* u = *cx; */
1844: /* *cx = dum; */
1845: /* dum = fu; */
1846: /* fu = *fc; */
1847: /* *fc =dum; */
1848: /* } else { /\* end *\/ */
1849: /* #ifdef DEBUG */
1850: /* printf("mnbrak3 fu < fc \n"); */
1851: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1852: /* #endif */
1853: /* dum=u; /\* Shifting c and u *\/ */
1854: /* u = *cx; */
1855: /* *cx = dum; */
1856: /* dum = fu; */
1857: /* fu = *fc; */
1858: /* *fc =dum; */
1859: /* } */
1.224 brouard 1860: #ifdef DEBUGMNBRAK
1861: double A, fparabu;
1862: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1863: fparabu= *fa - A*(*ax-u)*(*ax-u);
1864: 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);
1865: 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 1866: #endif
1.191 brouard 1867: dum=u; /* Shifting c and u */
1868: u = *cx;
1869: *cx = dum;
1870: dum = fu;
1871: fu = *fc;
1872: *fc =dum;
1.183 brouard 1873: #endif
1.162 brouard 1874: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1875: #ifdef DEBUG
1.224 brouard 1876: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1877: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1878: #endif
1.126 brouard 1879: fu=(*func)(u);
1880: if (fu < *fc) {
1.183 brouard 1881: #ifdef DEBUG
1.224 brouard 1882: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1883: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1884: #endif
1885: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1886: SHFT(*fb,*fc,fu,(*func)(u))
1887: #ifdef DEBUG
1888: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1889: #endif
1890: }
1.162 brouard 1891: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1892: #ifdef DEBUG
1.224 brouard 1893: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1894: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1895: #endif
1.126 brouard 1896: u=ulim;
1897: fu=(*func)(u);
1.183 brouard 1898: } else { /* u could be left to b (if r > q parabola has a maximum) */
1899: #ifdef DEBUG
1.224 brouard 1900: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1901: 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 1902: #endif
1.126 brouard 1903: u=(*cx)+GOLD*(*cx-*bx);
1904: fu=(*func)(u);
1.224 brouard 1905: #ifdef DEBUG
1906: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1907: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1908: #endif
1.183 brouard 1909: } /* end tests */
1.126 brouard 1910: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1911: SHFT(*fa,*fb,*fc,fu)
1912: #ifdef DEBUG
1.224 brouard 1913: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1914: 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 1915: #endif
1916: } /* 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 1917: }
1918:
1919: /*************** linmin ************************/
1.162 brouard 1920: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1921: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1922: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1923: the value of func at the returned location p . This is actually all accomplished by calling the
1924: routines mnbrak and brent .*/
1.126 brouard 1925: int ncom;
1926: double *pcom,*xicom;
1927: double (*nrfunc)(double []);
1928:
1.224 brouard 1929: #ifdef LINMINORIGINAL
1.126 brouard 1930: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1931: #else
1932: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1933: #endif
1.126 brouard 1934: {
1935: double brent(double ax, double bx, double cx,
1936: double (*f)(double), double tol, double *xmin);
1937: double f1dim(double x);
1938: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1939: double *fc, double (*func)(double));
1940: int j;
1941: double xx,xmin,bx,ax;
1942: double fx,fb,fa;
1.187 brouard 1943:
1.203 brouard 1944: #ifdef LINMINORIGINAL
1945: #else
1946: double scale=10., axs, xxs; /* Scale added for infinity */
1947: #endif
1948:
1.126 brouard 1949: ncom=n;
1950: pcom=vector(1,n);
1951: xicom=vector(1,n);
1952: nrfunc=func;
1953: for (j=1;j<=n;j++) {
1954: pcom[j]=p[j];
1.202 brouard 1955: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1956: }
1.187 brouard 1957:
1.203 brouard 1958: #ifdef LINMINORIGINAL
1959: xx=1.;
1960: #else
1961: axs=0.0;
1962: xxs=1.;
1963: do{
1964: xx= xxs;
1965: #endif
1.187 brouard 1966: ax=0.;
1967: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
1968: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
1969: /* 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)) */
1970: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
1971: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
1972: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
1973: /* 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 1974: #ifdef LINMINORIGINAL
1975: #else
1976: if (fx != fx){
1.224 brouard 1977: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
1978: printf("|");
1979: fprintf(ficlog,"|");
1.203 brouard 1980: #ifdef DEBUGLINMIN
1.224 brouard 1981: 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 1982: #endif
1983: }
1.224 brouard 1984: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 1985: #endif
1986:
1.191 brouard 1987: #ifdef DEBUGLINMIN
1988: 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 1989: 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 1990: #endif
1.224 brouard 1991: #ifdef LINMINORIGINAL
1992: #else
1993: if(fb == fx){ /* Flat function in the direction */
1994: xmin=xx;
1995: *flat=1;
1996: }else{
1997: *flat=0;
1998: #endif
1999: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2000: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2001: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2002: /* fmin = f(p[j] + xmin * xi[j]) */
2003: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2004: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2005: #ifdef DEBUG
1.224 brouard 2006: 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);
2007: 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);
2008: #endif
2009: #ifdef LINMINORIGINAL
2010: #else
2011: }
1.126 brouard 2012: #endif
1.191 brouard 2013: #ifdef DEBUGLINMIN
2014: printf("linmin end ");
1.202 brouard 2015: fprintf(ficlog,"linmin end ");
1.191 brouard 2016: #endif
1.126 brouard 2017: for (j=1;j<=n;j++) {
1.203 brouard 2018: #ifdef LINMINORIGINAL
2019: xi[j] *= xmin;
2020: #else
2021: #ifdef DEBUGLINMIN
2022: if(xxs <1.0)
2023: printf(" before xi[%d]=%12.8f", j,xi[j]);
2024: #endif
2025: 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) */
2026: #ifdef DEBUGLINMIN
2027: if(xxs <1.0)
2028: 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 );
2029: #endif
2030: #endif
1.187 brouard 2031: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2032: }
1.191 brouard 2033: #ifdef DEBUGLINMIN
1.203 brouard 2034: printf("\n");
1.191 brouard 2035: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2036: 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 2037: for (j=1;j<=n;j++) {
1.202 brouard 2038: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2039: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2040: if(j % ncovmodel == 0){
1.191 brouard 2041: printf("\n");
1.202 brouard 2042: fprintf(ficlog,"\n");
2043: }
1.191 brouard 2044: }
1.203 brouard 2045: #else
1.191 brouard 2046: #endif
1.126 brouard 2047: free_vector(xicom,1,n);
2048: free_vector(pcom,1,n);
2049: }
2050:
2051:
2052: /*************** powell ************************/
1.162 brouard 2053: /*
2054: Minimization of a function func of n variables. Input consists of an initial starting point
2055: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2056: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2057: such that failure to decrease by more than this amount on one iteration signals doneness. On
2058: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2059: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2060: */
1.224 brouard 2061: #ifdef LINMINORIGINAL
2062: #else
2063: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2064: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2065: #endif
1.126 brouard 2066: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2067: double (*func)(double []))
2068: {
1.224 brouard 2069: #ifdef LINMINORIGINAL
2070: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2071: double (*func)(double []));
1.224 brouard 2072: #else
1.241 ! brouard 2073: void linmin(double p[], double xi[], int n, double *fret,
! 2074: double (*func)(double []),int *flat);
1.224 brouard 2075: #endif
1.239 brouard 2076: int i,ibig,j,jk,k;
1.126 brouard 2077: double del,t,*pt,*ptt,*xit;
1.181 brouard 2078: double directest;
1.126 brouard 2079: double fp,fptt;
2080: double *xits;
2081: int niterf, itmp;
1.224 brouard 2082: #ifdef LINMINORIGINAL
2083: #else
2084:
2085: flatdir=ivector(1,n);
2086: for (j=1;j<=n;j++) flatdir[j]=0;
2087: #endif
1.126 brouard 2088:
2089: pt=vector(1,n);
2090: ptt=vector(1,n);
2091: xit=vector(1,n);
2092: xits=vector(1,n);
2093: *fret=(*func)(p);
2094: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2095: rcurr_time = time(NULL);
1.126 brouard 2096: for (*iter=1;;++(*iter)) {
1.187 brouard 2097: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2098: ibig=0;
2099: del=0.0;
1.157 brouard 2100: rlast_time=rcurr_time;
2101: /* (void) gettimeofday(&curr_time,&tzp); */
2102: rcurr_time = time(NULL);
2103: curr_time = *localtime(&rcurr_time);
2104: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2105: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2106: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2107: for (i=1;i<=n;i++) {
1.126 brouard 2108: fprintf(ficrespow," %.12lf", p[i]);
2109: }
1.239 brouard 2110: fprintf(ficrespow,"\n");fflush(ficrespow);
2111: printf("\n#model= 1 + age ");
2112: fprintf(ficlog,"\n#model= 1 + age ");
2113: if(nagesqr==1){
1.241 ! brouard 2114: printf(" + age*age ");
! 2115: fprintf(ficlog," + age*age ");
1.239 brouard 2116: }
2117: for(j=1;j <=ncovmodel-2;j++){
2118: if(Typevar[j]==0) {
2119: printf(" + V%d ",Tvar[j]);
2120: fprintf(ficlog," + V%d ",Tvar[j]);
2121: }else if(Typevar[j]==1) {
2122: printf(" + V%d*age ",Tvar[j]);
2123: fprintf(ficlog," + V%d*age ",Tvar[j]);
2124: }else if(Typevar[j]==2) {
2125: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2126: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2127: }
2128: }
1.126 brouard 2129: printf("\n");
1.239 brouard 2130: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2131: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2132: fprintf(ficlog,"\n");
1.239 brouard 2133: for(i=1,jk=1; i <=nlstate; i++){
2134: for(k=1; k <=(nlstate+ndeath); k++){
2135: if (k != i) {
2136: printf("%d%d ",i,k);
2137: fprintf(ficlog,"%d%d ",i,k);
2138: for(j=1; j <=ncovmodel; j++){
2139: printf("%12.7f ",p[jk]);
2140: fprintf(ficlog,"%12.7f ",p[jk]);
2141: jk++;
2142: }
2143: printf("\n");
2144: fprintf(ficlog,"\n");
2145: }
2146: }
2147: }
1.241 ! brouard 2148: if(*iter <=3 && *iter >1){
1.157 brouard 2149: tml = *localtime(&rcurr_time);
2150: strcpy(strcurr,asctime(&tml));
2151: rforecast_time=rcurr_time;
1.126 brouard 2152: itmp = strlen(strcurr);
2153: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 ! brouard 2154: strcurr[itmp-1]='\0';
1.162 brouard 2155: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2156: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2157: for(niterf=10;niterf<=30;niterf+=10){
1.241 ! brouard 2158: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
! 2159: forecast_time = *localtime(&rforecast_time);
! 2160: strcpy(strfor,asctime(&forecast_time));
! 2161: itmp = strlen(strfor);
! 2162: if(strfor[itmp-1]=='\n')
! 2163: strfor[itmp-1]='\0';
! 2164: 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);
! 2165: 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 2166: }
2167: }
1.187 brouard 2168: for (i=1;i<=n;i++) { /* For each direction i */
2169: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2170: fptt=(*fret);
2171: #ifdef DEBUG
1.203 brouard 2172: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2173: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2174: #endif
1.203 brouard 2175: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2176: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2177: #ifdef LINMINORIGINAL
1.188 brouard 2178: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2179: #else
2180: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2181: flatdir[i]=flat; /* Function is vanishing in that direction i */
2182: #endif
2183: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2184: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2185: /* because that direction will be replaced unless the gain del is small */
2186: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2187: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2188: /* with the new direction. */
2189: del=fabs(fptt-(*fret));
2190: ibig=i;
1.126 brouard 2191: }
2192: #ifdef DEBUG
2193: printf("%d %.12e",i,(*fret));
2194: fprintf(ficlog,"%d %.12e",i,(*fret));
2195: for (j=1;j<=n;j++) {
1.224 brouard 2196: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2197: printf(" x(%d)=%.12e",j,xit[j]);
2198: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2199: }
2200: for(j=1;j<=n;j++) {
1.225 brouard 2201: printf(" p(%d)=%.12e",j,p[j]);
2202: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2203: }
2204: printf("\n");
2205: fprintf(ficlog,"\n");
2206: #endif
1.187 brouard 2207: } /* end loop on each direction i */
2208: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2209: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2210: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2211: for(j=1;j<=n;j++) {
1.225 brouard 2212: if(flatdir[j] >0){
2213: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2214: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2215: }
2216: /* printf("\n"); */
2217: /* fprintf(ficlog,"\n"); */
2218: }
1.182 brouard 2219: if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /* Did we reach enough precision? */
1.188 brouard 2220: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2221: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2222: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2223: /* decreased of more than 3.84 */
2224: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2225: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2226: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2227:
1.188 brouard 2228: /* Starting the program with initial values given by a former maximization will simply change */
2229: /* the scales of the directions and the directions, because the are reset to canonical directions */
2230: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2231: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2232: #ifdef DEBUG
2233: int k[2],l;
2234: k[0]=1;
2235: k[1]=-1;
2236: printf("Max: %.12e",(*func)(p));
2237: fprintf(ficlog,"Max: %.12e",(*func)(p));
2238: for (j=1;j<=n;j++) {
2239: printf(" %.12e",p[j]);
2240: fprintf(ficlog," %.12e",p[j]);
2241: }
2242: printf("\n");
2243: fprintf(ficlog,"\n");
2244: for(l=0;l<=1;l++) {
2245: for (j=1;j<=n;j++) {
2246: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2247: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2248: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2249: }
2250: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2251: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2252: }
2253: #endif
2254:
1.224 brouard 2255: #ifdef LINMINORIGINAL
2256: #else
2257: free_ivector(flatdir,1,n);
2258: #endif
1.126 brouard 2259: free_vector(xit,1,n);
2260: free_vector(xits,1,n);
2261: free_vector(ptt,1,n);
2262: free_vector(pt,1,n);
2263: return;
1.192 brouard 2264: } /* enough precision */
1.240 brouard 2265: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2266: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2267: ptt[j]=2.0*p[j]-pt[j];
2268: xit[j]=p[j]-pt[j];
2269: pt[j]=p[j];
2270: }
1.181 brouard 2271: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2272: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2273: if (*iter <=4) {
1.225 brouard 2274: #else
2275: #endif
1.224 brouard 2276: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2277: #else
1.161 brouard 2278: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2279: #endif
1.162 brouard 2280: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2281: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2282: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2283: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2284: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2285: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2286: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2287: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2288: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2289: /* Even if f3 <f1, directest can be negative and t >0 */
2290: /* mu² and del² are equal when f3=f1 */
2291: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2292: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2293: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2294: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2295: #ifdef NRCORIGINAL
2296: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2297: #else
2298: 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 2299: t= t- del*SQR(fp-fptt);
1.183 brouard 2300: #endif
1.202 brouard 2301: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2302: #ifdef DEBUG
1.181 brouard 2303: 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);
2304: 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 2305: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2306: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2307: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2308: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2309: 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);
2310: 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);
2311: #endif
1.183 brouard 2312: #ifdef POWELLORIGINAL
2313: if (t < 0.0) { /* Then we use it for new direction */
2314: #else
1.182 brouard 2315: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2316: 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 2317: 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 2318: 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 2319: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2320: }
1.181 brouard 2321: if (directest < 0.0) { /* Then we use it for new direction */
2322: #endif
1.191 brouard 2323: #ifdef DEBUGLINMIN
1.234 brouard 2324: printf("Before linmin in direction P%d-P0\n",n);
2325: for (j=1;j<=n;j++) {
2326: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2327: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2328: if(j % ncovmodel == 0){
2329: printf("\n");
2330: fprintf(ficlog,"\n");
2331: }
2332: }
1.224 brouard 2333: #endif
2334: #ifdef LINMINORIGINAL
1.234 brouard 2335: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2336: #else
1.234 brouard 2337: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2338: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2339: #endif
1.234 brouard 2340:
1.191 brouard 2341: #ifdef DEBUGLINMIN
1.234 brouard 2342: for (j=1;j<=n;j++) {
2343: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2344: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2345: if(j % ncovmodel == 0){
2346: printf("\n");
2347: fprintf(ficlog,"\n");
2348: }
2349: }
1.224 brouard 2350: #endif
1.234 brouard 2351: for (j=1;j<=n;j++) {
2352: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2353: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2354: }
1.224 brouard 2355: #ifdef LINMINORIGINAL
2356: #else
1.234 brouard 2357: for (j=1, flatd=0;j<=n;j++) {
2358: if(flatdir[j]>0)
2359: flatd++;
2360: }
2361: if(flatd >0){
2362: printf("%d flat directions\n",flatd);
2363: fprintf(ficlog,"%d flat directions\n",flatd);
2364: for (j=1;j<=n;j++) {
2365: if(flatdir[j]>0){
2366: printf("%d ",j);
2367: fprintf(ficlog,"%d ",j);
2368: }
2369: }
2370: printf("\n");
2371: fprintf(ficlog,"\n");
2372: }
1.191 brouard 2373: #endif
1.234 brouard 2374: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2375: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2376:
1.126 brouard 2377: #ifdef DEBUG
1.234 brouard 2378: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2379: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2380: for(j=1;j<=n;j++){
2381: printf(" %lf",xit[j]);
2382: fprintf(ficlog," %lf",xit[j]);
2383: }
2384: printf("\n");
2385: fprintf(ficlog,"\n");
1.126 brouard 2386: #endif
1.192 brouard 2387: } /* end of t or directest negative */
1.224 brouard 2388: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2389: #else
1.234 brouard 2390: } /* end if (fptt < fp) */
1.192 brouard 2391: #endif
1.225 brouard 2392: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2393: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2394: #else
1.224 brouard 2395: #endif
1.234 brouard 2396: } /* loop iteration */
1.126 brouard 2397: }
1.234 brouard 2398:
1.126 brouard 2399: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2400:
1.235 brouard 2401: 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 2402: {
1.235 brouard 2403: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2404: (and selected quantitative values in nres)
2405: by left multiplying the unit
1.234 brouard 2406: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2407: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2408: /* Wx is row vector: population in state 1, population in state 2, population dead */
2409: /* or prevalence in state 1, prevalence in state 2, 0 */
2410: /* newm is the matrix after multiplications, its rows are identical at a factor */
2411: /* Initial matrix pimij */
2412: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2413: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2414: /* 0, 0 , 1} */
2415: /*
2416: * and after some iteration: */
2417: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2418: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2419: /* 0, 0 , 1} */
2420: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2421: /* {0.51571254859325999, 0.4842874514067399, */
2422: /* 0.51326036147820708, 0.48673963852179264} */
2423: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2424:
1.126 brouard 2425: int i, ii,j,k;
1.209 brouard 2426: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2427: /* double **matprod2(); */ /* test */
1.218 brouard 2428: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2429: double **newm;
1.209 brouard 2430: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2431: int ncvloop=0;
1.169 brouard 2432:
1.209 brouard 2433: min=vector(1,nlstate);
2434: max=vector(1,nlstate);
2435: meandiff=vector(1,nlstate);
2436:
1.218 brouard 2437: /* Starting with matrix unity */
1.126 brouard 2438: for (ii=1;ii<=nlstate+ndeath;ii++)
2439: for (j=1;j<=nlstate+ndeath;j++){
2440: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2441: }
1.169 brouard 2442:
2443: cov[1]=1.;
2444:
2445: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2446: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2447: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2448: ncvloop++;
1.126 brouard 2449: newm=savm;
2450: /* Covariates have to be included here again */
1.138 brouard 2451: cov[2]=agefin;
1.187 brouard 2452: if(nagesqr==1)
2453: cov[3]= agefin*agefin;;
1.234 brouard 2454: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2455: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2456: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2457: /* 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 2458: }
2459: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2460: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2461: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2462: /* 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 2463: }
1.237 brouard 2464: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2465: if(Dummy[Tvar[Tage[k]]]){
2466: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2467: } else{
1.235 brouard 2468: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2469: }
1.235 brouard 2470: /* 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 2471: }
1.237 brouard 2472: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2473: /* 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 2474: if(Dummy[Tvard[k][1]==0]){
2475: if(Dummy[Tvard[k][2]==0]){
2476: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2477: }else{
2478: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2479: }
2480: }else{
2481: if(Dummy[Tvard[k][2]==0]){
2482: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2483: }else{
2484: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2485: }
2486: }
1.234 brouard 2487: }
1.138 brouard 2488: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2489: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2490: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2491: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2492: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2493: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2494: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2495:
1.126 brouard 2496: savm=oldm;
2497: oldm=newm;
1.209 brouard 2498:
2499: for(j=1; j<=nlstate; j++){
2500: max[j]=0.;
2501: min[j]=1.;
2502: }
2503: for(i=1;i<=nlstate;i++){
2504: sumnew=0;
2505: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2506: for(j=1; j<=nlstate; j++){
2507: prlim[i][j]= newm[i][j]/(1-sumnew);
2508: max[j]=FMAX(max[j],prlim[i][j]);
2509: min[j]=FMIN(min[j],prlim[i][j]);
2510: }
2511: }
2512:
1.126 brouard 2513: maxmax=0.;
1.209 brouard 2514: for(j=1; j<=nlstate; j++){
2515: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2516: maxmax=FMAX(maxmax,meandiff[j]);
2517: /* 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 2518: } /* j loop */
1.203 brouard 2519: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2520: /* 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 2521: if(maxmax < ftolpl){
1.209 brouard 2522: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2523: free_vector(min,1,nlstate);
2524: free_vector(max,1,nlstate);
2525: free_vector(meandiff,1,nlstate);
1.126 brouard 2526: return prlim;
2527: }
1.169 brouard 2528: } /* age loop */
1.208 brouard 2529: /* After some age loop it doesn't converge */
1.209 brouard 2530: 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 2531: 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 2532: /* 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); */
2533: free_vector(min,1,nlstate);
2534: free_vector(max,1,nlstate);
2535: free_vector(meandiff,1,nlstate);
1.208 brouard 2536:
1.169 brouard 2537: return prlim; /* should not reach here */
1.126 brouard 2538: }
2539:
1.217 brouard 2540:
2541: /**** Back Prevalence limit (stable or period prevalence) ****************/
2542:
1.218 brouard 2543: /* 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) */
2544: /* 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) */
2545: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij)
1.217 brouard 2546: {
1.218 brouard 2547: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2548: matrix by transitions matrix until convergence is reached with precision ftolpl */
2549: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2550: /* Wx is row vector: population in state 1, population in state 2, population dead */
2551: /* or prevalence in state 1, prevalence in state 2, 0 */
2552: /* newm is the matrix after multiplications, its rows are identical at a factor */
2553: /* Initial matrix pimij */
2554: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2555: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2556: /* 0, 0 , 1} */
2557: /*
2558: * and after some iteration: */
2559: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2560: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2561: /* 0, 0 , 1} */
2562: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2563: /* {0.51571254859325999, 0.4842874514067399, */
2564: /* 0.51326036147820708, 0.48673963852179264} */
2565: /* If we start from prlim again, prlim tends to a constant matrix */
2566:
2567: int i, ii,j,k;
2568: double *min, *max, *meandiff, maxmax,sumnew=0.;
2569: /* double **matprod2(); */ /* test */
2570: double **out, cov[NCOVMAX+1], **bmij();
2571: double **newm;
1.218 brouard 2572: double **dnewm, **doldm, **dsavm; /* for use */
2573: double **oldm, **savm; /* for use */
2574:
1.217 brouard 2575: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2576: int ncvloop=0;
2577:
2578: min=vector(1,nlstate);
2579: max=vector(1,nlstate);
2580: meandiff=vector(1,nlstate);
2581:
1.218 brouard 2582: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2583: oldm=oldms; savm=savms;
2584:
2585: /* Starting with matrix unity */
2586: for (ii=1;ii<=nlstate+ndeath;ii++)
2587: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2588: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2589: }
2590:
2591: cov[1]=1.;
2592:
2593: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2594: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2595: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2596: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2597: ncvloop++;
1.218 brouard 2598: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2599: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2600: /* Covariates have to be included here again */
2601: cov[2]=agefin;
2602: if(nagesqr==1)
2603: cov[3]= agefin*agefin;;
2604: for (k=1; k<=cptcovn;k++) {
2605: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
2606: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
2607: /* 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])]); */
2608: }
2609: for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2];
2610: for (k=1; k<=cptcovprod;k++) /* Useless */
2611: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
2612: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2613:
2614: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2615: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2616: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2617: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2618: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2619: /* ij should be linked to the correct index of cov */
2620: /* age and covariate values ij are in 'cov', but we need to pass
2621: * ij for the observed prevalence at age and status and covariate
2622: * number: prevacurrent[(int)agefin][ii][ij]
2623: */
2624: /* 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 *\/ */
2625: /* 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 *\/ */
2626: 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 2627: savm=oldm;
2628: oldm=newm;
2629: for(j=1; j<=nlstate; j++){
2630: max[j]=0.;
2631: min[j]=1.;
2632: }
2633: for(j=1; j<=nlstate; j++){
2634: for(i=1;i<=nlstate;i++){
1.234 brouard 2635: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2636: bprlim[i][j]= newm[i][j];
2637: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2638: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2639: }
2640: }
1.218 brouard 2641:
1.217 brouard 2642: maxmax=0.;
2643: for(i=1; i<=nlstate; i++){
2644: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2645: maxmax=FMAX(maxmax,meandiff[i]);
2646: /* 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); */
2647: } /* j loop */
2648: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2649: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2650: if(maxmax < ftolpl){
1.220 brouard 2651: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2652: free_vector(min,1,nlstate);
2653: free_vector(max,1,nlstate);
2654: free_vector(meandiff,1,nlstate);
2655: return bprlim;
2656: }
2657: } /* age loop */
2658: /* After some age loop it doesn't converge */
2659: printf("Warning: the back stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. \n\
2660: 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);
2661: /* 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); */
2662: free_vector(min,1,nlstate);
2663: free_vector(max,1,nlstate);
2664: free_vector(meandiff,1,nlstate);
2665:
2666: return bprlim; /* should not reach here */
2667: }
2668:
1.126 brouard 2669: /*************** transition probabilities ***************/
2670:
2671: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2672: {
1.138 brouard 2673: /* According to parameters values stored in x and the covariate's values stored in cov,
2674: computes the probability to be observed in state j being in state i by appying the
2675: model to the ncovmodel covariates (including constant and age).
2676: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2677: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2678: ncth covariate in the global vector x is given by the formula:
2679: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2680: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2681: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2682: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2683: Outputs ps[i][j] the probability to be observed in j being in j according to
2684: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2685: */
2686: double s1, lnpijopii;
1.126 brouard 2687: /*double t34;*/
1.164 brouard 2688: int i,j, nc, ii, jj;
1.126 brouard 2689:
1.223 brouard 2690: for(i=1; i<= nlstate; i++){
2691: for(j=1; j<i;j++){
2692: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2693: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2694: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2695: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2696: }
2697: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2698: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2699: }
2700: for(j=i+1; j<=nlstate+ndeath;j++){
2701: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2702: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2703: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2704: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2705: }
2706: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2707: }
2708: }
1.218 brouard 2709:
1.223 brouard 2710: for(i=1; i<= nlstate; i++){
2711: s1=0;
2712: for(j=1; j<i; j++){
2713: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2714: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2715: }
2716: for(j=i+1; j<=nlstate+ndeath; j++){
2717: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2718: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2719: }
2720: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2721: ps[i][i]=1./(s1+1.);
2722: /* Computing other pijs */
2723: for(j=1; j<i; j++)
2724: ps[i][j]= exp(ps[i][j])*ps[i][i];
2725: for(j=i+1; j<=nlstate+ndeath; j++)
2726: ps[i][j]= exp(ps[i][j])*ps[i][i];
2727: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2728: } /* end i */
1.218 brouard 2729:
1.223 brouard 2730: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2731: for(jj=1; jj<= nlstate+ndeath; jj++){
2732: ps[ii][jj]=0;
2733: ps[ii][ii]=1;
2734: }
2735: }
1.218 brouard 2736:
2737:
1.223 brouard 2738: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2739: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2740: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2741: /* } */
2742: /* printf("\n "); */
2743: /* } */
2744: /* printf("\n ");printf("%lf ",cov[2]);*/
2745: /*
2746: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2747: goto end;*/
1.223 brouard 2748: return ps;
1.126 brouard 2749: }
2750:
1.218 brouard 2751: /*************** backward transition probabilities ***************/
2752:
2753: /* 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 ) */
2754: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2755: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2756: {
1.222 brouard 2757: /* Computes the backward probability at age agefin and covariate ij
2758: * and returns in **ps as well as **bmij.
2759: */
1.218 brouard 2760: int i, ii, j,k;
1.222 brouard 2761:
2762: double **out, **pmij();
2763: double sumnew=0.;
1.218 brouard 2764: double agefin;
1.222 brouard 2765:
2766: double **dnewm, **dsavm, **doldm;
2767: double **bbmij;
2768:
1.218 brouard 2769: doldm=ddoldms; /* global pointers */
1.222 brouard 2770: dnewm=ddnewms;
2771: dsavm=ddsavms;
2772:
2773: agefin=cov[2];
2774: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2775: the observed prevalence (with this covariate ij) */
2776: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2777: /* We do have the matrix Px in savm and we need pij */
2778: for (j=1;j<=nlstate+ndeath;j++){
2779: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2780: for (ii=1;ii<=nlstate;ii++){
2781: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2782: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2783: for (ii=1;ii<=nlstate+ndeath;ii++){
2784: if(sumnew >= 1.e-10){
2785: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2786: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2787: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2788: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2789: /* }else */
2790: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2791: }else{
2792: 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);
2793: }
2794: } /*End ii */
2795: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2796: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2797: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2798: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2799: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2800: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2801: /* left Product of this matrix by diag matrix of prevalences (savm) */
2802: for (j=1;j<=nlstate+ndeath;j++){
2803: for (ii=1;ii<=nlstate+ndeath;ii++){
2804: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2805: }
2806: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2807: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2808: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2809: /* end bmij */
2810: return ps;
1.218 brouard 2811: }
1.217 brouard 2812: /*************** transition probabilities ***************/
2813:
1.218 brouard 2814: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2815: {
2816: /* According to parameters values stored in x and the covariate's values stored in cov,
2817: computes the probability to be observed in state j being in state i by appying the
2818: model to the ncovmodel covariates (including constant and age).
2819: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2820: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2821: ncth covariate in the global vector x is given by the formula:
2822: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2823: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2824: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2825: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2826: Outputs ps[i][j] the probability to be observed in j being in j according to
2827: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2828: */
2829: double s1, lnpijopii;
2830: /*double t34;*/
2831: int i,j, nc, ii, jj;
2832:
1.234 brouard 2833: for(i=1; i<= nlstate; i++){
2834: for(j=1; j<i;j++){
2835: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2836: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2837: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2838: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2839: }
2840: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2841: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2842: }
2843: for(j=i+1; j<=nlstate+ndeath;j++){
2844: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2845: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2846: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2847: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2848: }
2849: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2850: }
2851: }
2852:
2853: for(i=1; i<= nlstate; i++){
2854: s1=0;
2855: for(j=1; j<i; j++){
2856: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2857: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2858: }
2859: for(j=i+1; j<=nlstate+ndeath; j++){
2860: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2861: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2862: }
2863: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2864: ps[i][i]=1./(s1+1.);
2865: /* Computing other pijs */
2866: for(j=1; j<i; j++)
2867: ps[i][j]= exp(ps[i][j])*ps[i][i];
2868: for(j=i+1; j<=nlstate+ndeath; j++)
2869: ps[i][j]= exp(ps[i][j])*ps[i][i];
2870: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2871: } /* end i */
2872:
2873: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2874: for(jj=1; jj<= nlstate+ndeath; jj++){
2875: ps[ii][jj]=0;
2876: ps[ii][ii]=1;
2877: }
2878: }
2879: /* Added for backcast */ /* Transposed matrix too */
2880: for(jj=1; jj<= nlstate+ndeath; jj++){
2881: s1=0.;
2882: for(ii=1; ii<= nlstate+ndeath; ii++){
2883: s1+=ps[ii][jj];
2884: }
2885: for(ii=1; ii<= nlstate; ii++){
2886: ps[ii][jj]=ps[ii][jj]/s1;
2887: }
2888: }
2889: /* Transposition */
2890: for(jj=1; jj<= nlstate+ndeath; jj++){
2891: for(ii=jj; ii<= nlstate+ndeath; ii++){
2892: s1=ps[ii][jj];
2893: ps[ii][jj]=ps[jj][ii];
2894: ps[jj][ii]=s1;
2895: }
2896: }
2897: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2898: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2899: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2900: /* } */
2901: /* printf("\n "); */
2902: /* } */
2903: /* printf("\n ");printf("%lf ",cov[2]);*/
2904: /*
2905: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2906: goto end;*/
2907: return ps;
1.217 brouard 2908: }
2909:
2910:
1.126 brouard 2911: /**************** Product of 2 matrices ******************/
2912:
1.145 brouard 2913: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2914: {
2915: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
2916: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
2917: /* in, b, out are matrice of pointers which should have been initialized
2918: before: only the contents of out is modified. The function returns
2919: a pointer to pointers identical to out */
1.145 brouard 2920: int i, j, k;
1.126 brouard 2921: for(i=nrl; i<= nrh; i++)
1.145 brouard 2922: for(k=ncolol; k<=ncoloh; k++){
2923: out[i][k]=0.;
2924: for(j=ncl; j<=nch; j++)
2925: out[i][k] +=in[i][j]*b[j][k];
2926: }
1.126 brouard 2927: return out;
2928: }
2929:
2930:
2931: /************* Higher Matrix Product ***************/
2932:
1.235 brouard 2933: 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 2934: {
1.218 brouard 2935: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 2936: 'nhstepm*hstepm*stepm' months (i.e. until
2937: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
2938: nhstepm*hstepm matrices.
2939: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
2940: (typically every 2 years instead of every month which is too big
2941: for the memory).
2942: Model is determined by parameters x and covariates have to be
2943: included manually here.
2944:
2945: */
2946:
2947: int i, j, d, h, k;
1.131 brouard 2948: double **out, cov[NCOVMAX+1];
1.126 brouard 2949: double **newm;
1.187 brouard 2950: double agexact;
1.214 brouard 2951: double agebegin, ageend;
1.126 brouard 2952:
2953: /* Hstepm could be zero and should return the unit matrix */
2954: for (i=1;i<=nlstate+ndeath;i++)
2955: for (j=1;j<=nlstate+ndeath;j++){
2956: oldm[i][j]=(i==j ? 1.0 : 0.0);
2957: po[i][j][0]=(i==j ? 1.0 : 0.0);
2958: }
2959: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2960: for(h=1; h <=nhstepm; h++){
2961: for(d=1; d <=hstepm; d++){
2962: newm=savm;
2963: /* Covariates have to be included here again */
2964: cov[1]=1.;
1.214 brouard 2965: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 2966: cov[2]=agexact;
2967: if(nagesqr==1)
1.227 brouard 2968: cov[3]= agexact*agexact;
1.235 brouard 2969: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2970: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2971: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2972: /* 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)); */
2973: }
2974: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2975: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2976: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2977: /* 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]); */
2978: }
2979: for (k=1; k<=cptcovage;k++){
2980: if(Dummy[Tvar[Tage[k]]]){
2981: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2982: } else{
2983: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2984: }
2985: /* 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]); */
2986: }
2987: for (k=1; k<=cptcovprod;k++){ /* */
2988: /* 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]); */
2989: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2990: }
2991: /* for (k=1; k<=cptcovn;k++) */
2992: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2993: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
2994: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
2995: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
2996: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 2997:
2998:
1.126 brouard 2999: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3000: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3001: /* right multiplication of oldm by the current matrix */
1.126 brouard 3002: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3003: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3004: /* if((int)age == 70){ */
3005: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3006: /* for(i=1; i<=nlstate+ndeath; i++) { */
3007: /* printf("%d pmmij ",i); */
3008: /* for(j=1;j<=nlstate+ndeath;j++) { */
3009: /* printf("%f ",pmmij[i][j]); */
3010: /* } */
3011: /* printf(" oldm "); */
3012: /* for(j=1;j<=nlstate+ndeath;j++) { */
3013: /* printf("%f ",oldm[i][j]); */
3014: /* } */
3015: /* printf("\n"); */
3016: /* } */
3017: /* } */
1.126 brouard 3018: savm=oldm;
3019: oldm=newm;
3020: }
3021: for(i=1; i<=nlstate+ndeath; i++)
3022: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3023: po[i][j][h]=newm[i][j];
3024: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3025: }
1.128 brouard 3026: /*printf("h=%d ",h);*/
1.126 brouard 3027: } /* end h */
1.218 brouard 3028: /* printf("\n H=%d \n",h); */
1.126 brouard 3029: return po;
3030: }
3031:
1.217 brouard 3032: /************* Higher Back Matrix Product ***************/
1.218 brouard 3033: /* 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 3034: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3035: {
1.218 brouard 3036: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3037: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3038: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3039: nhstepm*hstepm matrices.
3040: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3041: (typically every 2 years instead of every month which is too big
1.217 brouard 3042: for the memory).
1.218 brouard 3043: Model is determined by parameters x and covariates have to be
3044: included manually here.
1.217 brouard 3045:
1.222 brouard 3046: */
1.217 brouard 3047:
3048: int i, j, d, h, k;
3049: double **out, cov[NCOVMAX+1];
3050: double **newm;
3051: double agexact;
3052: double agebegin, ageend;
1.222 brouard 3053: double **oldm, **savm;
1.217 brouard 3054:
1.222 brouard 3055: oldm=oldms;savm=savms;
1.217 brouard 3056: /* Hstepm could be zero and should return the unit matrix */
3057: for (i=1;i<=nlstate+ndeath;i++)
3058: for (j=1;j<=nlstate+ndeath;j++){
3059: oldm[i][j]=(i==j ? 1.0 : 0.0);
3060: po[i][j][0]=(i==j ? 1.0 : 0.0);
3061: }
3062: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3063: for(h=1; h <=nhstepm; h++){
3064: for(d=1; d <=hstepm; d++){
3065: newm=savm;
3066: /* Covariates have to be included here again */
3067: cov[1]=1.;
3068: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3069: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3070: cov[2]=agexact;
3071: if(nagesqr==1)
1.222 brouard 3072: cov[3]= agexact*agexact;
1.218 brouard 3073: for (k=1; k<=cptcovn;k++)
1.222 brouard 3074: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3075: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3076: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3077: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3078: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3079: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3080: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3081: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3082: /* 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 3083:
3084:
1.217 brouard 3085: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3086: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3087: /* Careful transposed matrix */
1.222 brouard 3088: /* age is in cov[2] */
1.218 brouard 3089: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3090: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3091: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3092: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3093: /* if((int)age == 70){ */
3094: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3095: /* for(i=1; i<=nlstate+ndeath; i++) { */
3096: /* printf("%d pmmij ",i); */
3097: /* for(j=1;j<=nlstate+ndeath;j++) { */
3098: /* printf("%f ",pmmij[i][j]); */
3099: /* } */
3100: /* printf(" oldm "); */
3101: /* for(j=1;j<=nlstate+ndeath;j++) { */
3102: /* printf("%f ",oldm[i][j]); */
3103: /* } */
3104: /* printf("\n"); */
3105: /* } */
3106: /* } */
3107: savm=oldm;
3108: oldm=newm;
3109: }
3110: for(i=1; i<=nlstate+ndeath; i++)
3111: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3112: po[i][j][h]=newm[i][j];
3113: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3114: }
3115: /*printf("h=%d ",h);*/
3116: } /* end h */
1.222 brouard 3117: /* printf("\n H=%d \n",h); */
1.217 brouard 3118: return po;
3119: }
3120:
3121:
1.162 brouard 3122: #ifdef NLOPT
3123: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3124: double fret;
3125: double *xt;
3126: int j;
3127: myfunc_data *d2 = (myfunc_data *) pd;
3128: /* xt = (p1-1); */
3129: xt=vector(1,n);
3130: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3131:
3132: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3133: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3134: printf("Function = %.12lf ",fret);
3135: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3136: printf("\n");
3137: free_vector(xt,1,n);
3138: return fret;
3139: }
3140: #endif
1.126 brouard 3141:
3142: /*************** log-likelihood *************/
3143: double func( double *x)
3144: {
1.226 brouard 3145: int i, ii, j, k, mi, d, kk;
3146: int ioffset=0;
3147: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3148: double **out;
3149: double lli; /* Individual log likelihood */
3150: int s1, s2;
1.228 brouard 3151: 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 3152: double bbh, survp;
3153: long ipmx;
3154: double agexact;
3155: /*extern weight */
3156: /* We are differentiating ll according to initial status */
3157: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3158: /*for(i=1;i<imx;i++)
3159: printf(" %d\n",s[4][i]);
3160: */
1.162 brouard 3161:
1.226 brouard 3162: ++countcallfunc;
1.162 brouard 3163:
1.226 brouard 3164: cov[1]=1.;
1.126 brouard 3165:
1.226 brouard 3166: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3167: ioffset=0;
1.226 brouard 3168: if(mle==1){
3169: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3170: /* Computes the values of the ncovmodel covariates of the model
3171: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3172: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3173: to be observed in j being in i according to the model.
3174: */
3175: ioffset=2+nagesqr+cptcovage;
1.233 brouard 3176: /* Fixed */
1.234 brouard 3177: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3178: 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)*/
3179: }
1.226 brouard 3180: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3181: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3182: has been calculated etc */
3183: /* For an individual i, wav[i] gives the number of effective waves */
3184: /* We compute the contribution to Likelihood of each effective transition
3185: mw[mi][i] is real wave of the mi th effectve wave */
3186: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3187: s2=s[mw[mi+1][i]][i];
3188: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3189: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3190: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3191: */
3192: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3193: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
3194: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i];
3195: }
3196: for (ii=1;ii<=nlstate+ndeath;ii++)
3197: for (j=1;j<=nlstate+ndeath;j++){
3198: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3199: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3200: }
3201: for(d=0; d<dh[mi][i]; d++){
3202: newm=savm;
3203: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3204: cov[2]=agexact;
3205: if(nagesqr==1)
3206: cov[3]= agexact*agexact; /* Should be changed here */
3207: for (kk=1; kk<=cptcovage;kk++) {
3208: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
3209: }
3210: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3211: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3212: savm=oldm;
3213: oldm=newm;
3214: } /* end mult */
3215:
3216: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3217: /* But now since version 0.9 we anticipate for bias at large stepm.
3218: * If stepm is larger than one month (smallest stepm) and if the exact delay
3219: * (in months) between two waves is not a multiple of stepm, we rounded to
3220: * the nearest (and in case of equal distance, to the lowest) interval but now
3221: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3222: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3223: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3224: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3225: * -stepm/2 to stepm/2 .
3226: * For stepm=1 the results are the same as for previous versions of Imach.
3227: * For stepm > 1 the results are less biased than in previous versions.
3228: */
1.234 brouard 3229: s1=s[mw[mi][i]][i];
3230: s2=s[mw[mi+1][i]][i];
3231: bbh=(double)bh[mi][i]/(double)stepm;
3232: /* bias bh is positive if real duration
3233: * is higher than the multiple of stepm and negative otherwise.
3234: */
3235: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3236: if( s2 > nlstate){
3237: /* i.e. if s2 is a death state and if the date of death is known
3238: then the contribution to the likelihood is the probability to
3239: die between last step unit time and current step unit time,
3240: which is also equal to probability to die before dh
3241: minus probability to die before dh-stepm .
3242: In version up to 0.92 likelihood was computed
3243: as if date of death was unknown. Death was treated as any other
3244: health state: the date of the interview describes the actual state
3245: and not the date of a change in health state. The former idea was
3246: to consider that at each interview the state was recorded
3247: (healthy, disable or death) and IMaCh was corrected; but when we
3248: introduced the exact date of death then we should have modified
3249: the contribution of an exact death to the likelihood. This new
3250: contribution is smaller and very dependent of the step unit
3251: stepm. It is no more the probability to die between last interview
3252: and month of death but the probability to survive from last
3253: interview up to one month before death multiplied by the
3254: probability to die within a month. Thanks to Chris
3255: Jackson for correcting this bug. Former versions increased
3256: mortality artificially. The bad side is that we add another loop
3257: which slows down the processing. The difference can be up to 10%
3258: lower mortality.
3259: */
3260: /* If, at the beginning of the maximization mostly, the
3261: cumulative probability or probability to be dead is
3262: constant (ie = 1) over time d, the difference is equal to
3263: 0. out[s1][3] = savm[s1][3]: probability, being at state
3264: s1 at precedent wave, to be dead a month before current
3265: wave is equal to probability, being at state s1 at
3266: precedent wave, to be dead at mont of the current
3267: wave. Then the observed probability (that this person died)
3268: is null according to current estimated parameter. In fact,
3269: it should be very low but not zero otherwise the log go to
3270: infinity.
3271: */
1.183 brouard 3272: /* #ifdef INFINITYORIGINAL */
3273: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3274: /* #else */
3275: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3276: /* lli=log(mytinydouble); */
3277: /* else */
3278: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3279: /* #endif */
1.226 brouard 3280: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3281:
1.226 brouard 3282: } else if ( s2==-1 ) { /* alive */
3283: for (j=1,survp=0. ; j<=nlstate; j++)
3284: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3285: /*survp += out[s1][j]; */
3286: lli= log(survp);
3287: }
3288: else if (s2==-4) {
3289: for (j=3,survp=0. ; j<=nlstate; j++)
3290: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3291: lli= log(survp);
3292: }
3293: else if (s2==-5) {
3294: for (j=1,survp=0. ; j<=2; j++)
3295: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3296: lli= log(survp);
3297: }
3298: else{
3299: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3300: /* 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 */
3301: }
3302: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3303: /*if(lli ==000.0)*/
3304: /*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); */
3305: ipmx +=1;
3306: sw += weight[i];
3307: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3308: /* if (lli < log(mytinydouble)){ */
3309: /* 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); */
3310: /* 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]); */
3311: /* } */
3312: } /* end of wave */
3313: } /* end of individual */
3314: } else if(mle==2){
3315: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3316: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3317: for(mi=1; mi<= wav[i]-1; mi++){
3318: for (ii=1;ii<=nlstate+ndeath;ii++)
3319: for (j=1;j<=nlstate+ndeath;j++){
3320: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3321: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3322: }
3323: for(d=0; d<=dh[mi][i]; d++){
3324: newm=savm;
3325: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3326: cov[2]=agexact;
3327: if(nagesqr==1)
3328: cov[3]= agexact*agexact;
3329: for (kk=1; kk<=cptcovage;kk++) {
3330: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3331: }
3332: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3333: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3334: savm=oldm;
3335: oldm=newm;
3336: } /* end mult */
3337:
3338: s1=s[mw[mi][i]][i];
3339: s2=s[mw[mi+1][i]][i];
3340: bbh=(double)bh[mi][i]/(double)stepm;
3341: 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 */
3342: ipmx +=1;
3343: sw += weight[i];
3344: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3345: } /* end of wave */
3346: } /* end of individual */
3347: } else if(mle==3){ /* exponential inter-extrapolation */
3348: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3349: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3350: for(mi=1; mi<= wav[i]-1; mi++){
3351: for (ii=1;ii<=nlstate+ndeath;ii++)
3352: for (j=1;j<=nlstate+ndeath;j++){
3353: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3354: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3355: }
3356: for(d=0; d<dh[mi][i]; d++){
3357: newm=savm;
3358: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3359: cov[2]=agexact;
3360: if(nagesqr==1)
3361: cov[3]= agexact*agexact;
3362: for (kk=1; kk<=cptcovage;kk++) {
3363: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3364: }
3365: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3366: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3367: savm=oldm;
3368: oldm=newm;
3369: } /* end mult */
3370:
3371: s1=s[mw[mi][i]][i];
3372: s2=s[mw[mi+1][i]][i];
3373: bbh=(double)bh[mi][i]/(double)stepm;
3374: 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 */
3375: ipmx +=1;
3376: sw += weight[i];
3377: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3378: } /* end of wave */
3379: } /* end of individual */
3380: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3381: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3382: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3383: for(mi=1; mi<= wav[i]-1; mi++){
3384: for (ii=1;ii<=nlstate+ndeath;ii++)
3385: for (j=1;j<=nlstate+ndeath;j++){
3386: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3387: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3388: }
3389: for(d=0; d<dh[mi][i]; d++){
3390: newm=savm;
3391: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3392: cov[2]=agexact;
3393: if(nagesqr==1)
3394: cov[3]= agexact*agexact;
3395: for (kk=1; kk<=cptcovage;kk++) {
3396: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3397: }
1.126 brouard 3398:
1.226 brouard 3399: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3400: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3401: savm=oldm;
3402: oldm=newm;
3403: } /* end mult */
3404:
3405: s1=s[mw[mi][i]][i];
3406: s2=s[mw[mi+1][i]][i];
3407: if( s2 > nlstate){
3408: lli=log(out[s1][s2] - savm[s1][s2]);
3409: } else if ( s2==-1 ) { /* alive */
3410: for (j=1,survp=0. ; j<=nlstate; j++)
3411: survp += out[s1][j];
3412: lli= log(survp);
3413: }else{
3414: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3415: }
3416: ipmx +=1;
3417: sw += weight[i];
3418: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3419: /* 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 3420: } /* end of wave */
3421: } /* end of individual */
3422: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3423: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3424: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3425: for(mi=1; mi<= wav[i]-1; mi++){
3426: for (ii=1;ii<=nlstate+ndeath;ii++)
3427: for (j=1;j<=nlstate+ndeath;j++){
3428: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3429: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3430: }
3431: for(d=0; d<dh[mi][i]; d++){
3432: newm=savm;
3433: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3434: cov[2]=agexact;
3435: if(nagesqr==1)
3436: cov[3]= agexact*agexact;
3437: for (kk=1; kk<=cptcovage;kk++) {
3438: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3439: }
1.126 brouard 3440:
1.226 brouard 3441: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3442: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3443: savm=oldm;
3444: oldm=newm;
3445: } /* end mult */
3446:
3447: s1=s[mw[mi][i]][i];
3448: s2=s[mw[mi+1][i]][i];
3449: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3450: ipmx +=1;
3451: sw += weight[i];
3452: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3453: /*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]);*/
3454: } /* end of wave */
3455: } /* end of individual */
3456: } /* End of if */
3457: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3458: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3459: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3460: return -l;
1.126 brouard 3461: }
3462:
3463: /*************** log-likelihood *************/
3464: double funcone( double *x)
3465: {
1.228 brouard 3466: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3467: int i, ii, j, k, mi, d, kk;
1.228 brouard 3468: int ioffset=0;
1.131 brouard 3469: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3470: double **out;
3471: double lli; /* Individual log likelihood */
3472: double llt;
3473: int s1, s2;
1.228 brouard 3474: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3475:
1.126 brouard 3476: double bbh, survp;
1.187 brouard 3477: double agexact;
1.214 brouard 3478: double agebegin, ageend;
1.126 brouard 3479: /*extern weight */
3480: /* We are differentiating ll according to initial status */
3481: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3482: /*for(i=1;i<imx;i++)
3483: printf(" %d\n",s[4][i]);
3484: */
3485: cov[1]=1.;
3486:
3487: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3488: ioffset=0;
3489: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.225 brouard 3490: ioffset=2+nagesqr+cptcovage;
1.232 brouard 3491: /* Fixed */
1.224 brouard 3492: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3493: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3494: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3495: 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)*/
3496: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3497: /* cov[2+6]=covar[Tvar[6]][i]; */
3498: /* cov[2+6]=covar[2][i]; V2 */
3499: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3500: /* cov[2+7]=covar[Tvar[7]][i]; */
3501: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3502: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3503: /* cov[2+9]=covar[Tvar[9]][i]; */
3504: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3505: }
1.232 brouard 3506: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3507: /* 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?)*\/ */
3508: /* } */
1.231 brouard 3509: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3510: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3511: /* } */
1.225 brouard 3512:
1.233 brouard 3513:
3514: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3515: /* Wave varying (but not age varying) */
3516: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.233 brouard 3517: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i];
1.232 brouard 3518: }
3519: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.231 brouard 3520: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3521: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
1.232 brouard 3522: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3523: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
1.231 brouard 3524: /* 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 3525: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
3526: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3527: /* /\* 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]); *\/ */
3528: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
3529: /* } */
1.126 brouard 3530: for (ii=1;ii<=nlstate+ndeath;ii++)
1.231 brouard 3531: for (j=1;j<=nlstate+ndeath;j++){
3532: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3533: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3534: }
1.214 brouard 3535:
3536: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3537: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3538: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.231 brouard 3539: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3540: and mw[mi+1][i]. dh depends on stepm.*/
3541: newm=savm;
3542: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3543: cov[2]=agexact;
3544: if(nagesqr==1)
3545: cov[3]= agexact*agexact;
3546: for (kk=1; kk<=cptcovage;kk++) {
3547: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3548: }
3549: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3550: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3551: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3552: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3553: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3554: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3555: savm=oldm;
3556: oldm=newm;
1.126 brouard 3557: } /* end mult */
3558:
3559: s1=s[mw[mi][i]][i];
3560: s2=s[mw[mi+1][i]][i];
1.217 brouard 3561: /* if(s2==-1){ */
3562: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3563: /* /\* exit(1); *\/ */
3564: /* } */
1.126 brouard 3565: bbh=(double)bh[mi][i]/(double)stepm;
3566: /* bias is positive if real duration
3567: * is higher than the multiple of stepm and negative otherwise.
3568: */
3569: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.232 brouard 3570: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3571: } else if ( s2==-1 ) { /* alive */
1.232 brouard 3572: for (j=1,survp=0. ; j<=nlstate; j++)
3573: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3574: lli= log(survp);
1.126 brouard 3575: }else if (mle==1){
1.232 brouard 3576: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3577: } else if(mle==2){
1.232 brouard 3578: 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 3579: } else if(mle==3){ /* exponential inter-extrapolation */
1.232 brouard 3580: 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 3581: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.232 brouard 3582: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3583: } else{ /* mle=0 back to 1 */
1.232 brouard 3584: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3585: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3586: } /* End of if */
3587: ipmx +=1;
3588: sw += weight[i];
3589: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3590: /*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 3591: if(globpr){
1.232 brouard 3592: fprintf(ficresilk,"%9ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3593: %11.6f %11.6f %11.6f ", \
1.232 brouard 3594: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3595: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3596: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3597: llt +=ll[k]*gipmx/gsw;
3598: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3599: }
3600: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3601: }
1.232 brouard 3602: } /* end of wave */
3603: } /* end of individual */
3604: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3605: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3606: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3607: if(globpr==0){ /* First time we count the contributions and weights */
3608: gipmx=ipmx;
3609: gsw=sw;
3610: }
3611: return -l;
1.126 brouard 3612: }
3613:
3614:
3615: /*************** function likelione ***********/
3616: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3617: {
3618: /* This routine should help understanding what is done with
3619: the selection of individuals/waves and
3620: to check the exact contribution to the likelihood.
3621: Plotting could be done.
3622: */
3623: int k;
3624:
3625: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3626: strcpy(fileresilk,"ILK_");
1.202 brouard 3627: strcat(fileresilk,fileresu);
1.126 brouard 3628: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3629: printf("Problem with resultfile: %s\n", fileresilk);
3630: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3631: }
1.214 brouard 3632: 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");
3633: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3634: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3635: for(k=1; k<=nlstate; k++)
3636: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3637: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3638: }
3639:
3640: *fretone=(*funcone)(p);
3641: if(*globpri !=0){
3642: fclose(ficresilk);
1.205 brouard 3643: if (mle ==0)
3644: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3645: else if(mle >=1)
3646: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3647: 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 3648:
1.208 brouard 3649:
3650: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3651: 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 3652: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3653: }
1.207 brouard 3654: 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 3655: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3656: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3657: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3658: fflush(fichtm);
1.205 brouard 3659: }
1.126 brouard 3660: return;
3661: }
3662:
3663:
3664: /*********** Maximum Likelihood Estimation ***************/
3665:
3666: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3667: {
1.165 brouard 3668: int i,j, iter=0;
1.126 brouard 3669: double **xi;
3670: double fret;
3671: double fretone; /* Only one call to likelihood */
3672: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3673:
3674: #ifdef NLOPT
3675: int creturn;
3676: nlopt_opt opt;
3677: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3678: double *lb;
3679: double minf; /* the minimum objective value, upon return */
3680: double * p1; /* Shifted parameters from 0 instead of 1 */
3681: myfunc_data dinst, *d = &dinst;
3682: #endif
3683:
3684:
1.126 brouard 3685: xi=matrix(1,npar,1,npar);
3686: for (i=1;i<=npar;i++)
3687: for (j=1;j<=npar;j++)
3688: xi[i][j]=(i==j ? 1.0 : 0.0);
3689: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3690: strcpy(filerespow,"POW_");
1.126 brouard 3691: strcat(filerespow,fileres);
3692: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3693: printf("Problem with resultfile: %s\n", filerespow);
3694: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3695: }
3696: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3697: for (i=1;i<=nlstate;i++)
3698: for(j=1;j<=nlstate+ndeath;j++)
3699: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3700: fprintf(ficrespow,"\n");
1.162 brouard 3701: #ifdef POWELL
1.126 brouard 3702: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3703: #endif
1.126 brouard 3704:
1.162 brouard 3705: #ifdef NLOPT
3706: #ifdef NEWUOA
3707: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3708: #else
3709: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3710: #endif
3711: lb=vector(0,npar-1);
3712: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3713: nlopt_set_lower_bounds(opt, lb);
3714: nlopt_set_initial_step1(opt, 0.1);
3715:
3716: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3717: d->function = func;
3718: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3719: nlopt_set_min_objective(opt, myfunc, d);
3720: nlopt_set_xtol_rel(opt, ftol);
3721: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3722: printf("nlopt failed! %d\n",creturn);
3723: }
3724: else {
3725: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3726: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3727: iter=1; /* not equal */
3728: }
3729: nlopt_destroy(opt);
3730: #endif
1.126 brouard 3731: free_matrix(xi,1,npar,1,npar);
3732: fclose(ficrespow);
1.203 brouard 3733: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3734: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3735: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3736:
3737: }
3738:
3739: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3740: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3741: {
3742: double **a,**y,*x,pd;
1.203 brouard 3743: /* double **hess; */
1.164 brouard 3744: int i, j;
1.126 brouard 3745: int *indx;
3746:
3747: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3748: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3749: void lubksb(double **a, int npar, int *indx, double b[]) ;
3750: void ludcmp(double **a, int npar, int *indx, double *d) ;
3751: double gompertz(double p[]);
1.203 brouard 3752: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3753:
3754: printf("\nCalculation of the hessian matrix. Wait...\n");
3755: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3756: for (i=1;i<=npar;i++){
1.203 brouard 3757: printf("%d-",i);fflush(stdout);
3758: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3759:
3760: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3761:
3762: /* printf(" %f ",p[i]);
3763: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3764: }
3765:
3766: for (i=1;i<=npar;i++) {
3767: for (j=1;j<=npar;j++) {
3768: if (j>i) {
1.203 brouard 3769: printf(".%d-%d",i,j);fflush(stdout);
3770: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3771: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3772:
3773: hess[j][i]=hess[i][j];
3774: /*printf(" %lf ",hess[i][j]);*/
3775: }
3776: }
3777: }
3778: printf("\n");
3779: fprintf(ficlog,"\n");
3780:
3781: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3782: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3783:
3784: a=matrix(1,npar,1,npar);
3785: y=matrix(1,npar,1,npar);
3786: x=vector(1,npar);
3787: indx=ivector(1,npar);
3788: for (i=1;i<=npar;i++)
3789: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3790: ludcmp(a,npar,indx,&pd);
3791:
3792: for (j=1;j<=npar;j++) {
3793: for (i=1;i<=npar;i++) x[i]=0;
3794: x[j]=1;
3795: lubksb(a,npar,indx,x);
3796: for (i=1;i<=npar;i++){
3797: matcov[i][j]=x[i];
3798: }
3799: }
3800:
3801: printf("\n#Hessian matrix#\n");
3802: fprintf(ficlog,"\n#Hessian matrix#\n");
3803: for (i=1;i<=npar;i++) {
3804: for (j=1;j<=npar;j++) {
1.203 brouard 3805: printf("%.6e ",hess[i][j]);
3806: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3807: }
3808: printf("\n");
3809: fprintf(ficlog,"\n");
3810: }
3811:
1.203 brouard 3812: /* printf("\n#Covariance matrix#\n"); */
3813: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3814: /* for (i=1;i<=npar;i++) { */
3815: /* for (j=1;j<=npar;j++) { */
3816: /* printf("%.6e ",matcov[i][j]); */
3817: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3818: /* } */
3819: /* printf("\n"); */
3820: /* fprintf(ficlog,"\n"); */
3821: /* } */
3822:
1.126 brouard 3823: /* Recompute Inverse */
1.203 brouard 3824: /* for (i=1;i<=npar;i++) */
3825: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3826: /* ludcmp(a,npar,indx,&pd); */
3827:
3828: /* printf("\n#Hessian matrix recomputed#\n"); */
3829:
3830: /* for (j=1;j<=npar;j++) { */
3831: /* for (i=1;i<=npar;i++) x[i]=0; */
3832: /* x[j]=1; */
3833: /* lubksb(a,npar,indx,x); */
3834: /* for (i=1;i<=npar;i++){ */
3835: /* y[i][j]=x[i]; */
3836: /* printf("%.3e ",y[i][j]); */
3837: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3838: /* } */
3839: /* printf("\n"); */
3840: /* fprintf(ficlog,"\n"); */
3841: /* } */
3842:
3843: /* Verifying the inverse matrix */
3844: #ifdef DEBUGHESS
3845: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3846:
1.203 brouard 3847: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3848: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3849:
3850: for (j=1;j<=npar;j++) {
3851: for (i=1;i<=npar;i++){
1.203 brouard 3852: printf("%.2f ",y[i][j]);
3853: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3854: }
3855: printf("\n");
3856: fprintf(ficlog,"\n");
3857: }
1.203 brouard 3858: #endif
1.126 brouard 3859:
3860: free_matrix(a,1,npar,1,npar);
3861: free_matrix(y,1,npar,1,npar);
3862: free_vector(x,1,npar);
3863: free_ivector(indx,1,npar);
1.203 brouard 3864: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3865:
3866:
3867: }
3868:
3869: /*************** hessian matrix ****************/
3870: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3871: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3872: int i;
3873: int l=1, lmax=20;
1.203 brouard 3874: double k1,k2, res, fx;
1.132 brouard 3875: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3876: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3877: int k=0,kmax=10;
3878: double l1;
3879:
3880: fx=func(x);
3881: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3882: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3883: l1=pow(10,l);
3884: delts=delt;
3885: for(k=1 ; k <kmax; k=k+1){
3886: delt = delta*(l1*k);
3887: p2[theta]=x[theta] +delt;
1.145 brouard 3888: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3889: p2[theta]=x[theta]-delt;
3890: k2=func(p2)-fx;
3891: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3892: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3893:
1.203 brouard 3894: #ifdef DEBUGHESSII
1.126 brouard 3895: 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);
3896: 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);
3897: #endif
3898: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3899: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3900: k=kmax;
3901: }
3902: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3903: k=kmax; l=lmax*10;
1.126 brouard 3904: }
3905: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
3906: delts=delt;
3907: }
1.203 brouard 3908: } /* End loop k */
1.126 brouard 3909: }
3910: delti[theta]=delts;
3911: return res;
3912:
3913: }
3914:
1.203 brouard 3915: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 3916: {
3917: int i;
1.164 brouard 3918: int l=1, lmax=20;
1.126 brouard 3919: double k1,k2,k3,k4,res,fx;
1.132 brouard 3920: double p2[MAXPARM+1];
1.203 brouard 3921: int k, kmax=1;
3922: double v1, v2, cv12, lc1, lc2;
1.208 brouard 3923:
3924: int firstime=0;
1.203 brouard 3925:
1.126 brouard 3926: fx=func(x);
1.203 brouard 3927: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 3928: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 3929: p2[thetai]=x[thetai]+delti[thetai]*k;
3930: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 3931: k1=func(p2)-fx;
3932:
1.203 brouard 3933: p2[thetai]=x[thetai]+delti[thetai]*k;
3934: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 3935: k2=func(p2)-fx;
3936:
1.203 brouard 3937: p2[thetai]=x[thetai]-delti[thetai]*k;
3938: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 3939: k3=func(p2)-fx;
3940:
1.203 brouard 3941: p2[thetai]=x[thetai]-delti[thetai]*k;
3942: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 3943: k4=func(p2)-fx;
1.203 brouard 3944: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
3945: if(k1*k2*k3*k4 <0.){
1.208 brouard 3946: firstime=1;
1.203 brouard 3947: kmax=kmax+10;
1.208 brouard 3948: }
3949: if(kmax >=10 || firstime ==1){
1.218 brouard 3950: printf("Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you may increase ftol=%.2e\n",thetai,thetaj, ftol);
3951: fprintf(ficlog,"Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you may increase ftol=%.2e\n",thetai,thetaj, ftol);
1.203 brouard 3952: 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);
3953: 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);
3954: }
3955: #ifdef DEBUGHESSIJ
3956: v1=hess[thetai][thetai];
3957: v2=hess[thetaj][thetaj];
3958: cv12=res;
3959: /* Computing eigen value of Hessian matrix */
3960: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
3961: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
3962: if ((lc2 <0) || (lc1 <0) ){
3963: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
3964: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
3965: 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);
3966: 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);
3967: }
1.126 brouard 3968: #endif
3969: }
3970: return res;
3971: }
3972:
1.203 brouard 3973: /* Not done yet: Was supposed to fix if not exactly at the maximum */
3974: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
3975: /* { */
3976: /* int i; */
3977: /* int l=1, lmax=20; */
3978: /* double k1,k2,k3,k4,res,fx; */
3979: /* double p2[MAXPARM+1]; */
3980: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
3981: /* int k=0,kmax=10; */
3982: /* double l1; */
3983:
3984: /* fx=func(x); */
3985: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
3986: /* l1=pow(10,l); */
3987: /* delts=delt; */
3988: /* for(k=1 ; k <kmax; k=k+1){ */
3989: /* delt = delti*(l1*k); */
3990: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
3991: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
3992: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
3993: /* k1=func(p2)-fx; */
3994:
3995: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
3996: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
3997: /* k2=func(p2)-fx; */
3998:
3999: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4000: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4001: /* k3=func(p2)-fx; */
4002:
4003: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4004: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4005: /* k4=func(p2)-fx; */
4006: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4007: /* #ifdef DEBUGHESSIJ */
4008: /* 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); */
4009: /* 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); */
4010: /* #endif */
4011: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4012: /* k=kmax; */
4013: /* } */
4014: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4015: /* k=kmax; l=lmax*10; */
4016: /* } */
4017: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4018: /* delts=delt; */
4019: /* } */
4020: /* } /\* End loop k *\/ */
4021: /* } */
4022: /* delti[theta]=delts; */
4023: /* return res; */
4024: /* } */
4025:
4026:
1.126 brouard 4027: /************** Inverse of matrix **************/
4028: void ludcmp(double **a, int n, int *indx, double *d)
4029: {
4030: int i,imax,j,k;
4031: double big,dum,sum,temp;
4032: double *vv;
4033:
4034: vv=vector(1,n);
4035: *d=1.0;
4036: for (i=1;i<=n;i++) {
4037: big=0.0;
4038: for (j=1;j<=n;j++)
4039: if ((temp=fabs(a[i][j])) > big) big=temp;
4040: if (big == 0.0) nrerror("Singular matrix in routine ludcmp");
4041: vv[i]=1.0/big;
4042: }
4043: for (j=1;j<=n;j++) {
4044: for (i=1;i<j;i++) {
4045: sum=a[i][j];
4046: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4047: a[i][j]=sum;
4048: }
4049: big=0.0;
4050: for (i=j;i<=n;i++) {
4051: sum=a[i][j];
4052: for (k=1;k<j;k++)
4053: sum -= a[i][k]*a[k][j];
4054: a[i][j]=sum;
4055: if ( (dum=vv[i]*fabs(sum)) >= big) {
4056: big=dum;
4057: imax=i;
4058: }
4059: }
4060: if (j != imax) {
4061: for (k=1;k<=n;k++) {
4062: dum=a[imax][k];
4063: a[imax][k]=a[j][k];
4064: a[j][k]=dum;
4065: }
4066: *d = -(*d);
4067: vv[imax]=vv[j];
4068: }
4069: indx[j]=imax;
4070: if (a[j][j] == 0.0) a[j][j]=TINY;
4071: if (j != n) {
4072: dum=1.0/(a[j][j]);
4073: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4074: }
4075: }
4076: free_vector(vv,1,n); /* Doesn't work */
4077: ;
4078: }
4079:
4080: void lubksb(double **a, int n, int *indx, double b[])
4081: {
4082: int i,ii=0,ip,j;
4083: double sum;
4084:
4085: for (i=1;i<=n;i++) {
4086: ip=indx[i];
4087: sum=b[ip];
4088: b[ip]=b[i];
4089: if (ii)
4090: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4091: else if (sum) ii=i;
4092: b[i]=sum;
4093: }
4094: for (i=n;i>=1;i--) {
4095: sum=b[i];
4096: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4097: b[i]=sum/a[i][i];
4098: }
4099: }
4100:
4101: void pstamp(FILE *fichier)
4102: {
1.196 brouard 4103: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4104: }
4105:
4106: /************ Frequencies ********************/
1.226 brouard 4107: void freqsummary(char fileres[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
4108: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4109: int firstpass, int lastpass, int stepm, int weightopt, char model[])
4110: { /* Some frequencies */
4111:
1.227 brouard 4112: int i, m, jk, j1, bool, z1,j, k, iv;
1.226 brouard 4113: int iind=0, iage=0;
4114: int mi; /* Effective wave */
4115: int first;
4116: double ***freq; /* Frequencies */
4117: double *meanq;
4118: double **meanqt;
4119: double *pp, **prop, *posprop, *pospropt;
4120: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4121: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4122: double agebegin, ageend;
4123:
4124: pp=vector(1,nlstate);
4125: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4126: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4127: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4128: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4129: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4130: meanqt=matrix(1,lastpass,1,nqtveff);
4131: strcpy(fileresp,"P_");
4132: strcat(fileresp,fileresu);
4133: /*strcat(fileresphtm,fileresu);*/
4134: if((ficresp=fopen(fileresp,"w"))==NULL) {
4135: printf("Problem with prevalence resultfile: %s\n", fileresp);
4136: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4137: exit(0);
4138: }
1.240 brouard 4139:
1.226 brouard 4140: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4141: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4142: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4143: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4144: fflush(ficlog);
4145: exit(70);
4146: }
4147: else{
4148: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4149: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4150: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4151: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4152: }
1.237 brouard 4153: 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 4154:
1.226 brouard 4155: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4156: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4157: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4158: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4159: fflush(ficlog);
4160: exit(70);
1.240 brouard 4161: } else{
1.226 brouard 4162: 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 4163: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4164: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4165: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4166: }
1.240 brouard 4167: 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);
4168:
1.226 brouard 4169: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4170: j1=0;
1.126 brouard 4171:
1.227 brouard 4172: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4173: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4174: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4175:
1.226 brouard 4176: first=1;
1.240 brouard 4177:
1.226 brouard 4178: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4179: reference=low_education V1=0,V2=0
4180: med_educ V1=1 V2=0,
4181: high_educ V1=0 V2=1
4182: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4183: */
1.240 brouard 4184:
1.227 brouard 4185: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on covariates combination in order of model, excluding quantitatives V4=0, V3=0 for example, fixed or varying covariates */
1.226 brouard 4186: posproptt=0.;
4187: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4188: scanf("%d", i);*/
4189: for (i=-5; i<=nlstate+ndeath; i++)
4190: for (jk=-5; jk<=nlstate+ndeath; jk++)
1.240 brouard 4191: for(m=iagemin; m <= iagemax+3; m++)
4192: freq[i][jk][m]=0;
4193:
1.226 brouard 4194: for (i=1; i<=nlstate; i++) {
4195: for(m=iagemin; m <= iagemax+3; m++)
1.240 brouard 4196: prop[i][m]=0;
1.226 brouard 4197: posprop[i]=0;
4198: pospropt[i]=0;
4199: }
1.227 brouard 4200: /* for (z1=1; z1<= nqfveff; z1++) { */
4201: /* meanq[z1]+=0.; */
4202: /* for(m=1;m<=lastpass;m++){ */
4203: /* meanqt[m][z1]=0.; */
4204: /* } */
4205: /* } */
1.240 brouard 4206:
1.226 brouard 4207: dateintsum=0;
4208: k2cpt=0;
1.227 brouard 4209: /* For that combination of covariate j1, we count and print the frequencies in one pass */
1.226 brouard 4210: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4211: bool=1;
1.227 brouard 4212: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.234 brouard 4213: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
1.227 brouard 4214: /* for (z1=1; z1<= nqfveff; z1++) { */
4215: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4216: /* } */
1.234 brouard 4217: for (z1=1; z1<=cptcoveff; z1++) {
4218: /* if(Tvaraff[z1] ==-20){ */
4219: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4220: /* }else if(Tvaraff[z1] ==-10){ */
4221: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4222: /* }else */
4223: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){
4224: /* Tests if this individual iind responded to j1 (V4=1 V3=0) */
4225: bool=0;
4226: /* 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",
4227: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4228: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4229: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4230: } /* Onlyf fixed */
4231: } /* end z1 */
4232: } /* cptcovn > 0 */
1.227 brouard 4233: } /* end any */
4234: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
1.234 brouard 4235: /* for(m=firstpass; m<=lastpass; m++){ */
4236: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4237: m=mw[mi][iind];
4238: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4239: for (z1=1; z1<=cptcoveff; z1++) {
4240: if( Fixed[Tmodelind[z1]]==1){
4241: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4242: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4243: bool=0;
4244: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4245: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4246: bool=0;
4247: }
4248: }
4249: }
4250: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4251: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4252: if(bool==1){
4253: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4254: and mw[mi+1][iind]. dh depends on stepm. */
4255: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4256: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4257: if(m >=firstpass && m <=lastpass){
4258: k2=anint[m][iind]+(mint[m][iind]/12.);
4259: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4260: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4261: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4262: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4263: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4264: if (m<lastpass) {
4265: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4266: /* 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]); */
4267: if(s[m][iind]==-1)
4268: 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.));
4269: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4270: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4271: 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 */
4272: }
4273: } /* end if between passes */
4274: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99)) {
4275: dateintsum=dateintsum+k2;
4276: k2cpt++;
4277: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
4278: }
4279: } /* end bool 2 */
4280: } /* end m */
1.226 brouard 4281: } /* end bool */
4282: } /* end iind = 1 to imx */
4283: /* prop[s][age] is feeded for any initial and valid live state as well as
4284: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
1.240 brouard 4285:
4286:
1.226 brouard 4287: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4288: pstamp(ficresp);
1.240 brouard 4289: if (cptcoveff>0){
1.226 brouard 4290: fprintf(ficresp, "\n#********** Variable ");
4291: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4292: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
1.240 brouard 4293: fprintf(ficlog, "\n#********** Variable ");
1.227 brouard 4294: for (z1=1; z1<=cptcoveff; z1++){
1.240 brouard 4295: if(DummyV[z1]){
4296: fprintf(ficresp, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4297: fprintf(ficresphtm, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4298: fprintf(ficresphtmfr, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4299: fprintf(ficlog, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4300: }else{
4301: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4302: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4303: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4304: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4305: }
1.226 brouard 4306: }
4307: fprintf(ficresp, "**********\n#");
4308: fprintf(ficresphtm, "**********</h3>\n");
4309: fprintf(ficresphtmfr, "**********</h3>\n");
4310: fprintf(ficlog, "**********\n");
4311: }
4312: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4313: for(i=1; i<=nlstate;i++) {
1.240 brouard 4314: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
1.226 brouard 4315: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4316: }
4317: fprintf(ficresp, "\n");
4318: fprintf(ficresphtm, "\n");
1.240 brouard 4319:
1.226 brouard 4320: /* Header of frequency table by age */
4321: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4322: fprintf(ficresphtmfr,"<th>Age</th> ");
4323: for(jk=-1; jk <=nlstate+ndeath; jk++){
4324: for(m=-1; m <=nlstate+ndeath; m++){
1.234 brouard 4325: if(jk!=0 && m!=0)
4326: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.226 brouard 4327: }
4328: }
4329: fprintf(ficresphtmfr, "\n");
1.240 brouard 4330:
1.226 brouard 4331: /* For each age */
4332: for(iage=iagemin; iage <= iagemax+3; iage++){
4333: fprintf(ficresphtm,"<tr>");
4334: if(iage==iagemax+1){
1.240 brouard 4335: fprintf(ficlog,"1");
4336: fprintf(ficresphtmfr,"<tr><th>0</th> ");
1.226 brouard 4337: }else if(iage==iagemax+2){
1.240 brouard 4338: fprintf(ficlog,"0");
4339: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
1.226 brouard 4340: }else if(iage==iagemax+3){
1.240 brouard 4341: fprintf(ficlog,"Total");
4342: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
1.226 brouard 4343: }else{
1.240 brouard 4344: if(first==1){
4345: first=0;
4346: printf("See log file for details...\n");
4347: }
4348: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4349: fprintf(ficlog,"Age %d", iage);
1.226 brouard 4350: }
4351: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4352: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4353: pp[jk] += freq[jk][m][iage];
1.226 brouard 4354: }
4355: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4356: for(m=-1, pos=0; m <=0 ; m++)
4357: pos += freq[jk][m][iage];
4358: if(pp[jk]>=1.e-10){
4359: if(first==1){
4360: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4361: }
4362: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4363: }else{
4364: if(first==1)
4365: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4366: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4367: }
1.226 brouard 4368: }
1.240 brouard 4369:
1.226 brouard 4370: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4371: /* posprop[jk]=0; */
4372: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4373: pp[jk] += freq[jk][m][iage];
1.226 brouard 4374: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
1.240 brouard 4375:
1.226 brouard 4376: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
1.240 brouard 4377: pos += pp[jk]; /* pos is the total number of transitions until this age */
4378: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4379: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4380: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
4381: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
1.226 brouard 4382: }
4383: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4384: if(pos>=1.e-5){
4385: if(first==1)
4386: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4387: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4388: }else{
4389: if(first==1)
4390: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4391: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4392: }
4393: if( iage <= iagemax){
4394: if(pos>=1.e-5){
4395: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4396: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4397: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4398: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4399: }
4400: else{
4401: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4402: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4403: }
4404: }
4405: pospropt[jk] +=posprop[jk];
1.226 brouard 4406: } /* end loop jk */
4407: /* pospropt=0.; */
4408: for(jk=-1; jk <=nlstate+ndeath; jk++){
1.240 brouard 4409: for(m=-1; m <=nlstate+ndeath; m++){
4410: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4411: if(first==1){
4412: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4413: }
4414: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4415: }
4416: if(jk!=0 && m!=0)
4417: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
4418: }
1.226 brouard 4419: } /* end loop jk */
4420: posproptt=0.;
4421: for(jk=1; jk <=nlstate; jk++){
1.240 brouard 4422: posproptt += pospropt[jk];
1.226 brouard 4423: }
4424: fprintf(ficresphtmfr,"</tr>\n ");
4425: if(iage <= iagemax){
1.240 brouard 4426: fprintf(ficresp,"\n");
4427: fprintf(ficresphtm,"</tr>\n");
1.226 brouard 4428: }
4429: if(first==1)
1.240 brouard 4430: printf("Others in log...\n");
1.226 brouard 4431: fprintf(ficlog,"\n");
4432: } /* end loop age iage */
4433: fprintf(ficresphtm,"<tr><th>Tot</th>");
4434: for(jk=1; jk <=nlstate ; jk++){
4435: if(posproptt < 1.e-5){
1.240 brouard 4436: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
1.226 brouard 4437: }else{
1.240 brouard 4438: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.226 brouard 4439: }
4440: }
4441: fprintf(ficresphtm,"</tr>\n");
4442: fprintf(ficresphtm,"</table>\n");
4443: fprintf(ficresphtmfr,"</table>\n");
4444: if(posproptt < 1.e-5){
4445: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4446: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4447: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4448: invalidvarcomb[j1]=1;
4449: }else{
4450: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4451: invalidvarcomb[j1]=0;
4452: }
4453: fprintf(ficresphtmfr,"</table>\n");
4454: } /* end selected combination of covariate j1 */
4455: dateintmean=dateintsum/k2cpt;
1.240 brouard 4456:
1.226 brouard 4457: fclose(ficresp);
4458: fclose(ficresphtm);
4459: fclose(ficresphtmfr);
4460: free_vector(meanq,1,nqfveff);
4461: free_matrix(meanqt,1,lastpass,1,nqtveff);
4462: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4463: free_vector(pospropt,1,nlstate);
4464: free_vector(posprop,1,nlstate);
4465: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4466: free_vector(pp,1,nlstate);
4467: /* End of freqsummary */
4468: }
1.126 brouard 4469:
4470: /************ Prevalence ********************/
1.227 brouard 4471: 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)
4472: {
4473: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4474: in each health status at the date of interview (if between dateprev1 and dateprev2).
4475: We still use firstpass and lastpass as another selection.
4476: */
1.126 brouard 4477:
1.227 brouard 4478: int i, m, jk, j1, bool, z1,j, iv;
4479: int mi; /* Effective wave */
4480: int iage;
4481: double agebegin, ageend;
4482:
4483: double **prop;
4484: double posprop;
4485: double y2; /* in fractional years */
4486: int iagemin, iagemax;
4487: int first; /** to stop verbosity which is redirected to log file */
4488:
4489: iagemin= (int) agemin;
4490: iagemax= (int) agemax;
4491: /*pp=vector(1,nlstate);*/
4492: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4493: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4494: j1=0;
1.222 brouard 4495:
1.227 brouard 4496: /*j=cptcoveff;*/
4497: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4498:
1.227 brouard 4499: first=1;
4500: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4501: for (i=1; i<=nlstate; i++)
4502: for(iage=iagemin-AGEMARGE; iage <= iagemax+3+AGEMARGE; iage++)
4503: prop[i][iage]=0.0;
4504: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4505: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4506: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4507:
4508: for (i=1; i<=imx; i++) { /* Each individual */
4509: bool=1;
4510: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4511: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4512: m=mw[mi][i];
4513: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4514: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4515: for (z1=1; z1<=cptcoveff; z1++){
4516: if( Fixed[Tmodelind[z1]]==1){
4517: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4518: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4519: bool=0;
4520: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4521: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4522: bool=0;
4523: }
4524: }
4525: if(bool==1){ /* Otherwise we skip that wave/person */
4526: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4527: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4528: if(m >=firstpass && m <=lastpass){
4529: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4530: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4531: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4532: if(agev[m][i]==1) agev[m][i]=iagemax+2;
4533: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+3+AGEMARGE){
4534: 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);
4535: exit(1);
4536: }
4537: if (s[m][i]>0 && s[m][i]<=nlstate) {
4538: /*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]]);*/
4539: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4540: prop[s[m][i]][iagemax+3] += weight[i];
4541: } /* end valid statuses */
4542: } /* end selection of dates */
4543: } /* end selection of waves */
4544: } /* end bool */
4545: } /* end wave */
4546: } /* end individual */
4547: for(i=iagemin; i <= iagemax+3; i++){
4548: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4549: posprop += prop[jk][i];
4550: }
4551:
4552: for(jk=1; jk <=nlstate ; jk++){
4553: if( i <= iagemax){
4554: if(posprop>=1.e-5){
4555: probs[i][jk][j1]= prop[jk][i]/posprop;
4556: } else{
4557: if(first==1){
4558: first=0;
4559: 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]);
4560: }
4561: }
4562: }
4563: }/* end jk */
4564: }/* end i */
1.222 brouard 4565: /*} *//* end i1 */
1.227 brouard 4566: } /* end j1 */
1.222 brouard 4567:
1.227 brouard 4568: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4569: /*free_vector(pp,1,nlstate);*/
4570: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4571: } /* End of prevalence */
1.126 brouard 4572:
4573: /************* Waves Concatenation ***************/
4574:
4575: 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)
4576: {
4577: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4578: Death is a valid wave (if date is known).
4579: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4580: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4581: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4582: */
1.126 brouard 4583:
1.224 brouard 4584: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4585: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4586: double sum=0., jmean=0.;*/
1.224 brouard 4587: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4588: int j, k=0,jk, ju, jl;
4589: double sum=0.;
4590: first=0;
1.214 brouard 4591: firstwo=0;
1.217 brouard 4592: firsthree=0;
1.218 brouard 4593: firstfour=0;
1.164 brouard 4594: jmin=100000;
1.126 brouard 4595: jmax=-1;
4596: jmean=0.;
1.224 brouard 4597:
4598: /* Treating live states */
1.214 brouard 4599: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4600: mi=0; /* First valid wave */
1.227 brouard 4601: mli=0; /* Last valid wave */
1.126 brouard 4602: m=firstpass;
1.214 brouard 4603: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4604: 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 */
4605: mli=m-1;/* mw[++mi][i]=m-1; */
4606: }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 */
4607: mw[++mi][i]=m;
4608: mli=m;
1.224 brouard 4609: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4610: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4611: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4612: }
1.227 brouard 4613: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4614: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4615: break;
1.224 brouard 4616: #else
1.227 brouard 4617: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4618: if(firsthree == 0){
4619: 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);
4620: firsthree=1;
4621: }
4622: 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);
4623: mw[++mi][i]=m;
4624: mli=m;
4625: }
4626: if(s[m][i]==-2){ /* Vital status is really unknown */
4627: nbwarn++;
4628: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4629: 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);
4630: 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);
4631: }
4632: break;
4633: }
4634: break;
1.224 brouard 4635: #endif
1.227 brouard 4636: }/* End m >= lastpass */
1.126 brouard 4637: }/* end while */
1.224 brouard 4638:
1.227 brouard 4639: /* 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 4640: /* After last pass */
1.224 brouard 4641: /* Treating death states */
1.214 brouard 4642: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4643: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4644: /* } */
1.126 brouard 4645: mi++; /* Death is another wave */
4646: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4647: /* Only death is a correct wave */
1.126 brouard 4648: mw[mi][i]=m;
1.224 brouard 4649: }
4650: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4651: 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 4652: /* m++; */
4653: /* mi++; */
4654: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4655: /* mw[mi][i]=m; */
1.218 brouard 4656: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4657: 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 */
4658: nbwarn++;
4659: if(firstfiv==0){
4660: 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 );
4661: firstfiv=1;
4662: }else{
4663: 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 );
4664: }
4665: }else{ /* Death occured afer last wave potential bias */
4666: nberr++;
4667: if(firstwo==0){
4668: 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 );
4669: firstwo=1;
4670: }
4671: 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 );
4672: }
1.218 brouard 4673: }else{ /* end date of interview is known */
1.227 brouard 4674: /* death is known but not confirmed by death status at any wave */
4675: if(firstfour==0){
4676: 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 );
4677: firstfour=1;
4678: }
4679: 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 4680: }
1.224 brouard 4681: } /* end if date of death is known */
4682: #endif
4683: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4684: /* wav[i]=mw[mi][i]; */
1.126 brouard 4685: if(mi==0){
4686: nbwarn++;
4687: if(first==0){
1.227 brouard 4688: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4689: first=1;
1.126 brouard 4690: }
4691: if(first==1){
1.227 brouard 4692: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 4693: }
4694: } /* end mi==0 */
4695: } /* End individuals */
1.214 brouard 4696: /* wav and mw are no more changed */
1.223 brouard 4697:
1.214 brouard 4698:
1.126 brouard 4699: for(i=1; i<=imx; i++){
4700: for(mi=1; mi<wav[i];mi++){
4701: if (stepm <=0)
1.227 brouard 4702: dh[mi][i]=1;
1.126 brouard 4703: else{
1.227 brouard 4704: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
4705: if (agedc[i] < 2*AGESUP) {
4706: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
4707: if(j==0) j=1; /* Survives at least one month after exam */
4708: else if(j<0){
4709: nberr++;
4710: 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]);
4711: j=1; /* Temporary Dangerous patch */
4712: 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);
4713: 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]);
4714: 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);
4715: }
4716: k=k+1;
4717: if (j >= jmax){
4718: jmax=j;
4719: ijmax=i;
4720: }
4721: if (j <= jmin){
4722: jmin=j;
4723: ijmin=i;
4724: }
4725: sum=sum+j;
4726: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
4727: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
4728: }
4729: }
4730: else{
4731: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 4732: /* 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 4733:
1.227 brouard 4734: k=k+1;
4735: if (j >= jmax) {
4736: jmax=j;
4737: ijmax=i;
4738: }
4739: else if (j <= jmin){
4740: jmin=j;
4741: ijmin=i;
4742: }
4743: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
4744: /*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]);*/
4745: if(j<0){
4746: nberr++;
4747: 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]);
4748: 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]);
4749: }
4750: sum=sum+j;
4751: }
4752: jk= j/stepm;
4753: jl= j -jk*stepm;
4754: ju= j -(jk+1)*stepm;
4755: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
4756: if(jl==0){
4757: dh[mi][i]=jk;
4758: bh[mi][i]=0;
4759: }else{ /* We want a negative bias in order to only have interpolation ie
4760: * to avoid the price of an extra matrix product in likelihood */
4761: dh[mi][i]=jk+1;
4762: bh[mi][i]=ju;
4763: }
4764: }else{
4765: if(jl <= -ju){
4766: dh[mi][i]=jk;
4767: bh[mi][i]=jl; /* bias is positive if real duration
4768: * is higher than the multiple of stepm and negative otherwise.
4769: */
4770: }
4771: else{
4772: dh[mi][i]=jk+1;
4773: bh[mi][i]=ju;
4774: }
4775: if(dh[mi][i]==0){
4776: dh[mi][i]=1; /* At least one step */
4777: bh[mi][i]=ju; /* At least one step */
4778: /* 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);*/
4779: }
4780: } /* end if mle */
1.126 brouard 4781: }
4782: } /* end wave */
4783: }
4784: jmean=sum/k;
4785: 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 4786: 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 4787: }
1.126 brouard 4788:
4789: /*********** Tricode ****************************/
1.220 brouard 4790: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.126 brouard 4791: {
1.144 brouard 4792: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
4793: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
1.169 brouard 4794: * Boring subroutine which should only output nbcode[Tvar[j]][k]
1.224 brouard 4795: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
4796: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
1.144 brouard 4797: */
1.130 brouard 4798:
1.145 brouard 4799: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
1.136 brouard 4800: int modmaxcovj=0; /* Modality max of covariates j */
1.145 brouard 4801: int cptcode=0; /* Modality max of covariates j */
4802: int modmincovj=0; /* Modality min of covariates j */
4803:
4804:
1.220 brouard 4805: /* cptcoveff=0; */
1.224 brouard 4806: /* *cptcov=0; */
1.126 brouard 4807:
1.144 brouard 4808: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 4809:
1.224 brouard 4810: /* Loop on covariates without age and products and no quantitative variable */
4811: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
1.227 brouard 4812: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
4813: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
4814: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
4815: switch(Fixed[k]) {
4816: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.231 brouard 4817: 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*/
4818: ij=(int)(covar[Tvar[k]][i]);
4819: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
4820: * If product of Vn*Vm, still boolean *:
4821: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
4822: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
4823: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
4824: modality of the nth covariate of individual i. */
4825: if (ij > modmaxcovj)
4826: modmaxcovj=ij;
4827: else if (ij < modmincovj)
4828: modmincovj=ij;
4829: if ((ij < -1) && (ij > NCOVMAX)){
4830: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
4831: exit(1);
4832: }else
4833: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
4834: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
4835: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
4836: /* getting the maximum value of the modality of the covariate
4837: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
4838: female ies 1, then modmaxcovj=1.
4839: */
4840: } /* end for loop on individuals i */
4841: printf(" Minimal and maximal values of %d th covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
4842: fprintf(ficlog," Minimal and maximal values of %d th covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
4843: cptcode=modmaxcovj;
4844: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
4845: /*for (i=0; i<=cptcode; i++) {*/
4846: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
4847: printf("Frequencies of covariates %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
4848: fprintf(ficlog, "Frequencies of covariates %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
4849: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
4850: if( j != -1){
4851: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
4852: covariate for which somebody answered excluding
4853: undefined. Usually 2: 0 and 1. */
4854: }
4855: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
4856: covariate for which somebody answered including
4857: undefined. Usually 3: -1, 0 and 1. */
4858: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
4859: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
4860: } /* Ndum[-1] number of undefined modalities */
4861:
4862: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
4863: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
4864: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
4865: /* modmincovj=3; modmaxcovj = 7; */
4866: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
4867: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
4868: /* defining two dummy variables: variables V1_1 and V1_2.*/
4869: /* nbcode[Tvar[j]][ij]=k; */
4870: /* nbcode[Tvar[j]][1]=0; */
4871: /* nbcode[Tvar[j]][2]=1; */
4872: /* nbcode[Tvar[j]][3]=2; */
4873: /* To be continued (not working yet). */
4874: ij=0; /* ij is similar to i but can jump over null modalities */
4875: 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*/
4876: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
4877: break;
4878: }
4879: ij++;
4880: 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*/
4881: cptcode = ij; /* New max modality for covar j */
4882: } /* end of loop on modality i=-1 to 1 or more */
4883: break;
1.227 brouard 4884: case 1: /* Testing on varying covariate, could be simple and
4885: * should look at waves or product of fixed *
4886: * varying. No time to test -1, assuming 0 and 1 only */
1.231 brouard 4887: ij=0;
4888: for(i=0; i<=1;i++){
4889: nbcode[Tvar[k]][++ij]=i;
4890: }
4891: break;
1.227 brouard 4892: default:
1.231 brouard 4893: break;
1.227 brouard 4894: } /* end switch */
4895: } /* end dummy test */
1.225 brouard 4896:
1.192 brouard 4897: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
4898: /* /\*recode from 0 *\/ */
4899: /* k is a modality. If we have model=V1+V1*sex */
4900: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
4901: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
4902: /* } */
4903: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
4904: /* if (ij > ncodemax[j]) { */
4905: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
4906: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
4907: /* break; */
4908: /* } */
4909: /* } /\* end of loop on modality k *\/ */
1.137 brouard 4910: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
4911:
1.225 brouard 4912: for (k=-1; k< maxncov; k++) Ndum[k]=0;
1.227 brouard 4913: /* Look at fixed dummy (single or product) covariates to check empty modalities */
1.187 brouard 4914: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
1.225 brouard 4915: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
1.227 brouard 4916: 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 */
4917: 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 */
4918: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
1.225 brouard 4919: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
4920:
4921: ij=0;
1.227 brouard 4922: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
4923: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.225 brouard 4924: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
1.227 brouard 4925: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
4926: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
4927: /* If product not in single variable we don't print results */
1.225 brouard 4928: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.230 brouard 4929: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
4930: 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*/
4931: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
1.231 brouard 4932: 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 */
1.227 brouard 4933: if(Fixed[k]!=0)
4934: anyvaryingduminmodel=1;
1.231 brouard 4935: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
4936: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
4937: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
4938: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
4939: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
4940: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
1.227 brouard 4941: }
1.225 brouard 4942: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
4943: /* ij--; */
4944: /* cptcoveff=ij; /\*Number of total covariates*\/ */
4945: *cptcov=ij; /*Number of total real effective covariates: effective
1.231 brouard 4946: * because they can be excluded from the model and real
4947: * if in the model but excluded because missing values, but how to get k from ij?*/
1.227 brouard 4948: for(j=ij+1; j<= cptcovt; j++){
4949: Tvaraff[j]=0;
4950: Tmodelind[j]=0;
4951: }
1.228 brouard 4952: for(j=ntveff+1; j<= cptcovt; j++){
4953: TmodelInvind[j]=0;
4954: }
1.227 brouard 4955: /* To be sorted */
4956: ;
1.126 brouard 4957: }
4958:
1.145 brouard 4959:
1.126 brouard 4960: /*********** Health Expectancies ****************/
4961:
1.235 brouard 4962: 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 4963:
4964: {
4965: /* Health expectancies, no variances */
1.164 brouard 4966: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 4967: int nhstepma, nstepma; /* Decreasing with age */
4968: double age, agelim, hf;
4969: double ***p3mat;
4970: double eip;
4971:
1.238 brouard 4972: /* pstamp(ficreseij); */
1.126 brouard 4973: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
4974: fprintf(ficreseij,"# Age");
4975: for(i=1; i<=nlstate;i++){
4976: for(j=1; j<=nlstate;j++){
4977: fprintf(ficreseij," e%1d%1d ",i,j);
4978: }
4979: fprintf(ficreseij," e%1d. ",i);
4980: }
4981: fprintf(ficreseij,"\n");
4982:
4983:
4984: if(estepm < stepm){
4985: printf ("Problem %d lower than %d\n",estepm, stepm);
4986: }
4987: else hstepm=estepm;
4988: /* We compute the life expectancy from trapezoids spaced every estepm months
4989: * This is mainly to measure the difference between two models: for example
4990: * if stepm=24 months pijx are given only every 2 years and by summing them
4991: * we are calculating an estimate of the Life Expectancy assuming a linear
4992: * progression in between and thus overestimating or underestimating according
4993: * to the curvature of the survival function. If, for the same date, we
4994: * estimate the model with stepm=1 month, we can keep estepm to 24 months
4995: * to compare the new estimate of Life expectancy with the same linear
4996: * hypothesis. A more precise result, taking into account a more precise
4997: * curvature will be obtained if estepm is as small as stepm. */
4998:
4999: /* For example we decided to compute the life expectancy with the smallest unit */
5000: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5001: nhstepm is the number of hstepm from age to agelim
5002: nstepm is the number of stepm from age to agelin.
5003: Look at hpijx to understand the reason of that which relies in memory size
5004: and note for a fixed period like estepm months */
5005: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5006: survival function given by stepm (the optimization length). Unfortunately it
5007: means that if the survival funtion is printed only each two years of age and if
5008: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5009: results. So we changed our mind and took the option of the best precision.
5010: */
5011: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5012:
5013: agelim=AGESUP;
5014: /* If stepm=6 months */
5015: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5016: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5017:
5018: /* nhstepm age range expressed in number of stepm */
5019: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5020: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5021: /* if (stepm >= YEARM) hstepm=1;*/
5022: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5023: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5024:
5025: for (age=bage; age<=fage; age ++){
5026: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5027: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5028: /* if (stepm >= YEARM) hstepm=1;*/
5029: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5030:
5031: /* If stepm=6 months */
5032: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5033: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5034:
1.235 brouard 5035: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5036:
5037: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5038:
5039: printf("%d|",(int)age);fflush(stdout);
5040: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5041:
5042: /* Computing expectancies */
5043: for(i=1; i<=nlstate;i++)
5044: for(j=1; j<=nlstate;j++)
5045: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5046: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5047:
5048: /* 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]);*/
5049:
5050: }
5051:
5052: fprintf(ficreseij,"%3.0f",age );
5053: for(i=1; i<=nlstate;i++){
5054: eip=0;
5055: for(j=1; j<=nlstate;j++){
5056: eip +=eij[i][j][(int)age];
5057: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5058: }
5059: fprintf(ficreseij,"%9.4f", eip );
5060: }
5061: fprintf(ficreseij,"\n");
5062:
5063: }
5064: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5065: printf("\n");
5066: fprintf(ficlog,"\n");
5067:
5068: }
5069:
1.235 brouard 5070: 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 5071:
5072: {
5073: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5074: to initial status i, ei. .
1.126 brouard 5075: */
5076: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5077: int nhstepma, nstepma; /* Decreasing with age */
5078: double age, agelim, hf;
5079: double ***p3matp, ***p3matm, ***varhe;
5080: double **dnewm,**doldm;
5081: double *xp, *xm;
5082: double **gp, **gm;
5083: double ***gradg, ***trgradg;
5084: int theta;
5085:
5086: double eip, vip;
5087:
5088: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5089: xp=vector(1,npar);
5090: xm=vector(1,npar);
5091: dnewm=matrix(1,nlstate*nlstate,1,npar);
5092: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5093:
5094: pstamp(ficresstdeij);
5095: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5096: fprintf(ficresstdeij,"# Age");
5097: for(i=1; i<=nlstate;i++){
5098: for(j=1; j<=nlstate;j++)
5099: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5100: fprintf(ficresstdeij," e%1d. ",i);
5101: }
5102: fprintf(ficresstdeij,"\n");
5103:
5104: pstamp(ficrescveij);
5105: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5106: fprintf(ficrescveij,"# Age");
5107: for(i=1; i<=nlstate;i++)
5108: for(j=1; j<=nlstate;j++){
5109: cptj= (j-1)*nlstate+i;
5110: for(i2=1; i2<=nlstate;i2++)
5111: for(j2=1; j2<=nlstate;j2++){
5112: cptj2= (j2-1)*nlstate+i2;
5113: if(cptj2 <= cptj)
5114: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5115: }
5116: }
5117: fprintf(ficrescveij,"\n");
5118:
5119: if(estepm < stepm){
5120: printf ("Problem %d lower than %d\n",estepm, stepm);
5121: }
5122: else hstepm=estepm;
5123: /* We compute the life expectancy from trapezoids spaced every estepm months
5124: * This is mainly to measure the difference between two models: for example
5125: * if stepm=24 months pijx are given only every 2 years and by summing them
5126: * we are calculating an estimate of the Life Expectancy assuming a linear
5127: * progression in between and thus overestimating or underestimating according
5128: * to the curvature of the survival function. If, for the same date, we
5129: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5130: * to compare the new estimate of Life expectancy with the same linear
5131: * hypothesis. A more precise result, taking into account a more precise
5132: * curvature will be obtained if estepm is as small as stepm. */
5133:
5134: /* For example we decided to compute the life expectancy with the smallest unit */
5135: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5136: nhstepm is the number of hstepm from age to agelim
5137: nstepm is the number of stepm from age to agelin.
5138: Look at hpijx to understand the reason of that which relies in memory size
5139: and note for a fixed period like estepm months */
5140: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5141: survival function given by stepm (the optimization length). Unfortunately it
5142: means that if the survival funtion is printed only each two years of age and if
5143: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5144: results. So we changed our mind and took the option of the best precision.
5145: */
5146: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5147:
5148: /* If stepm=6 months */
5149: /* nhstepm age range expressed in number of stepm */
5150: agelim=AGESUP;
5151: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5152: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5153: /* if (stepm >= YEARM) hstepm=1;*/
5154: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5155:
5156: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5157: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5158: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5159: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5160: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5161: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5162:
5163: for (age=bage; age<=fage; age ++){
5164: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5165: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5166: /* if (stepm >= YEARM) hstepm=1;*/
5167: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5168:
1.126 brouard 5169: /* If stepm=6 months */
5170: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5171: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5172:
5173: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5174:
1.126 brouard 5175: /* Computing Variances of health expectancies */
5176: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5177: decrease memory allocation */
5178: for(theta=1; theta <=npar; theta++){
5179: for(i=1; i<=npar; i++){
1.222 brouard 5180: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5181: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5182: }
1.235 brouard 5183: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5184: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5185:
1.126 brouard 5186: for(j=1; j<= nlstate; j++){
1.222 brouard 5187: for(i=1; i<=nlstate; i++){
5188: for(h=0; h<=nhstepm-1; h++){
5189: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5190: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5191: }
5192: }
1.126 brouard 5193: }
1.218 brouard 5194:
1.126 brouard 5195: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5196: for(h=0; h<=nhstepm-1; h++){
5197: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5198: }
1.126 brouard 5199: }/* End theta */
5200:
5201:
5202: for(h=0; h<=nhstepm-1; h++)
5203: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5204: for(theta=1; theta <=npar; theta++)
5205: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5206:
1.218 brouard 5207:
1.222 brouard 5208: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5209: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5210: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5211:
1.222 brouard 5212: printf("%d|",(int)age);fflush(stdout);
5213: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5214: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5215: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5216: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5217: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5218: for(ij=1;ij<=nlstate*nlstate;ij++)
5219: for(ji=1;ji<=nlstate*nlstate;ji++)
5220: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5221: }
5222: }
1.218 brouard 5223:
1.126 brouard 5224: /* Computing expectancies */
1.235 brouard 5225: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5226: for(i=1; i<=nlstate;i++)
5227: for(j=1; j<=nlstate;j++)
1.222 brouard 5228: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5229: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5230:
1.222 brouard 5231: /* 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 5232:
1.222 brouard 5233: }
1.218 brouard 5234:
1.126 brouard 5235: fprintf(ficresstdeij,"%3.0f",age );
5236: for(i=1; i<=nlstate;i++){
5237: eip=0.;
5238: vip=0.;
5239: for(j=1; j<=nlstate;j++){
1.222 brouard 5240: eip += eij[i][j][(int)age];
5241: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5242: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5243: 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 5244: }
5245: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5246: }
5247: fprintf(ficresstdeij,"\n");
1.218 brouard 5248:
1.126 brouard 5249: fprintf(ficrescveij,"%3.0f",age );
5250: for(i=1; i<=nlstate;i++)
5251: for(j=1; j<=nlstate;j++){
1.222 brouard 5252: cptj= (j-1)*nlstate+i;
5253: for(i2=1; i2<=nlstate;i2++)
5254: for(j2=1; j2<=nlstate;j2++){
5255: cptj2= (j2-1)*nlstate+i2;
5256: if(cptj2 <= cptj)
5257: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5258: }
1.126 brouard 5259: }
5260: fprintf(ficrescveij,"\n");
1.218 brouard 5261:
1.126 brouard 5262: }
5263: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5264: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5265: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5266: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5267: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5268: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5269: printf("\n");
5270: fprintf(ficlog,"\n");
1.218 brouard 5271:
1.126 brouard 5272: free_vector(xm,1,npar);
5273: free_vector(xp,1,npar);
5274: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5275: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5276: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5277: }
1.218 brouard 5278:
1.126 brouard 5279: /************ Variance ******************/
1.235 brouard 5280: 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 5281: {
5282: /* Variance of health expectancies */
5283: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5284: /* double **newm;*/
5285: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5286:
5287: /* int movingaverage(); */
5288: double **dnewm,**doldm;
5289: double **dnewmp,**doldmp;
5290: int i, j, nhstepm, hstepm, h, nstepm ;
5291: int k;
5292: double *xp;
5293: double **gp, **gm; /* for var eij */
5294: double ***gradg, ***trgradg; /*for var eij */
5295: double **gradgp, **trgradgp; /* for var p point j */
5296: double *gpp, *gmp; /* for var p point j */
5297: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5298: double ***p3mat;
5299: double age,agelim, hf;
5300: /* double ***mobaverage; */
5301: int theta;
5302: char digit[4];
5303: char digitp[25];
5304:
5305: char fileresprobmorprev[FILENAMELENGTH];
5306:
5307: if(popbased==1){
5308: if(mobilav!=0)
5309: strcpy(digitp,"-POPULBASED-MOBILAV_");
5310: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5311: }
5312: else
5313: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5314:
1.218 brouard 5315: /* if (mobilav!=0) { */
5316: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5317: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5318: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5319: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5320: /* } */
5321: /* } */
5322:
5323: strcpy(fileresprobmorprev,"PRMORPREV-");
5324: sprintf(digit,"%-d",ij);
5325: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5326: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5327: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5328: strcat(fileresprobmorprev,fileresu);
5329: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5330: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5331: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5332: }
5333: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5334: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5335: pstamp(ficresprobmorprev);
5336: 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 5337: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5338: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5339: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5340: }
5341: for(j=1;j<=cptcoveff;j++)
5342: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5343: fprintf(ficresprobmorprev,"\n");
5344:
1.218 brouard 5345: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5346: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5347: fprintf(ficresprobmorprev," p.%-d SE",j);
5348: for(i=1; i<=nlstate;i++)
5349: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5350: }
5351: fprintf(ficresprobmorprev,"\n");
5352:
5353: fprintf(ficgp,"\n# Routine varevsij");
5354: fprintf(ficgp,"\nunset title \n");
5355: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5356: 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");
5357: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5358: /* } */
5359: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5360: pstamp(ficresvij);
5361: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5362: if(popbased==1)
5363: 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);
5364: else
5365: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5366: fprintf(ficresvij,"# Age");
5367: for(i=1; i<=nlstate;i++)
5368: for(j=1; j<=nlstate;j++)
5369: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5370: fprintf(ficresvij,"\n");
5371:
5372: xp=vector(1,npar);
5373: dnewm=matrix(1,nlstate,1,npar);
5374: doldm=matrix(1,nlstate,1,nlstate);
5375: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5376: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5377:
5378: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5379: gpp=vector(nlstate+1,nlstate+ndeath);
5380: gmp=vector(nlstate+1,nlstate+ndeath);
5381: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5382:
1.218 brouard 5383: if(estepm < stepm){
5384: printf ("Problem %d lower than %d\n",estepm, stepm);
5385: }
5386: else hstepm=estepm;
5387: /* For example we decided to compute the life expectancy with the smallest unit */
5388: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5389: nhstepm is the number of hstepm from age to agelim
5390: nstepm is the number of stepm from age to agelim.
5391: Look at function hpijx to understand why because of memory size limitations,
5392: we decided (b) to get a life expectancy respecting the most precise curvature of the
5393: survival function given by stepm (the optimization length). Unfortunately it
5394: means that if the survival funtion is printed every two years of age and if
5395: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5396: results. So we changed our mind and took the option of the best precision.
5397: */
5398: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5399: agelim = AGESUP;
5400: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5401: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5402: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5403: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5404: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5405: gp=matrix(0,nhstepm,1,nlstate);
5406: gm=matrix(0,nhstepm,1,nlstate);
5407:
5408:
5409: for(theta=1; theta <=npar; theta++){
5410: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5411: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5412: }
5413:
1.235 brouard 5414: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nresult);
1.218 brouard 5415:
5416: if (popbased==1) {
5417: if(mobilav ==0){
5418: for(i=1; i<=nlstate;i++)
5419: prlim[i][i]=probs[(int)age][i][ij];
5420: }else{ /* mobilav */
5421: for(i=1; i<=nlstate;i++)
5422: prlim[i][i]=mobaverage[(int)age][i][ij];
5423: }
5424: }
5425:
1.235 brouard 5426: 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 5427: for(j=1; j<= nlstate; j++){
5428: for(h=0; h<=nhstepm; h++){
5429: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5430: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5431: }
5432: }
5433: /* Next for computing probability of death (h=1 means
5434: computed over hstepm matrices product = hstepm*stepm months)
5435: as a weighted average of prlim.
5436: */
5437: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5438: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5439: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5440: }
5441: /* end probability of death */
5442:
5443: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5444: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5445:
1.235 brouard 5446: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nresult);
1.218 brouard 5447:
5448: if (popbased==1) {
5449: if(mobilav ==0){
5450: for(i=1; i<=nlstate;i++)
5451: prlim[i][i]=probs[(int)age][i][ij];
5452: }else{ /* mobilav */
5453: for(i=1; i<=nlstate;i++)
5454: prlim[i][i]=mobaverage[(int)age][i][ij];
5455: }
5456: }
5457:
1.235 brouard 5458: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5459:
5460: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5461: for(h=0; h<=nhstepm; h++){
5462: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5463: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5464: }
5465: }
5466: /* This for computing probability of death (h=1 means
5467: computed over hstepm matrices product = hstepm*stepm months)
5468: as a weighted average of prlim.
5469: */
5470: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5471: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5472: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5473: }
5474: /* end probability of death */
5475:
5476: for(j=1; j<= nlstate; j++) /* vareij */
5477: for(h=0; h<=nhstepm; h++){
5478: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5479: }
5480:
5481: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5482: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5483: }
5484:
5485: } /* End theta */
5486:
5487: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5488:
5489: for(h=0; h<=nhstepm; h++) /* veij */
5490: for(j=1; j<=nlstate;j++)
5491: for(theta=1; theta <=npar; theta++)
5492: trgradg[h][j][theta]=gradg[h][theta][j];
5493:
5494: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5495: for(theta=1; theta <=npar; theta++)
5496: trgradgp[j][theta]=gradgp[theta][j];
5497:
5498:
5499: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5500: for(i=1;i<=nlstate;i++)
5501: for(j=1;j<=nlstate;j++)
5502: vareij[i][j][(int)age] =0.;
5503:
5504: for(h=0;h<=nhstepm;h++){
5505: for(k=0;k<=nhstepm;k++){
5506: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5507: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5508: for(i=1;i<=nlstate;i++)
5509: for(j=1;j<=nlstate;j++)
5510: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5511: }
5512: }
5513:
5514: /* pptj */
5515: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5516: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5517: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5518: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5519: varppt[j][i]=doldmp[j][i];
5520: /* end ppptj */
5521: /* x centered again */
5522:
1.235 brouard 5523: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nresult);
1.218 brouard 5524:
5525: if (popbased==1) {
5526: if(mobilav ==0){
5527: for(i=1; i<=nlstate;i++)
5528: prlim[i][i]=probs[(int)age][i][ij];
5529: }else{ /* mobilav */
5530: for(i=1; i<=nlstate;i++)
5531: prlim[i][i]=mobaverage[(int)age][i][ij];
5532: }
5533: }
5534:
5535: /* This for computing probability of death (h=1 means
5536: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5537: as a weighted average of prlim.
5538: */
1.235 brouard 5539: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5540: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5541: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5542: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5543: }
5544: /* end probability of death */
5545:
5546: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5547: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5548: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5549: for(i=1; i<=nlstate;i++){
5550: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5551: }
5552: }
5553: fprintf(ficresprobmorprev,"\n");
5554:
5555: fprintf(ficresvij,"%.0f ",age );
5556: for(i=1; i<=nlstate;i++)
5557: for(j=1; j<=nlstate;j++){
5558: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5559: }
5560: fprintf(ficresvij,"\n");
5561: free_matrix(gp,0,nhstepm,1,nlstate);
5562: free_matrix(gm,0,nhstepm,1,nlstate);
5563: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5564: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5565: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5566: } /* End age */
5567: free_vector(gpp,nlstate+1,nlstate+ndeath);
5568: free_vector(gmp,nlstate+1,nlstate+ndeath);
5569: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5570: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5571: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5572: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5573: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5574: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5575: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5576: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5577: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5578: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5579: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5580: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5581: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5582: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5583: 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);
5584: /* 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 5585: */
1.218 brouard 5586: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5587: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5588:
1.218 brouard 5589: free_vector(xp,1,npar);
5590: free_matrix(doldm,1,nlstate,1,nlstate);
5591: free_matrix(dnewm,1,nlstate,1,npar);
5592: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5593: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5594: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5595: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5596: fclose(ficresprobmorprev);
5597: fflush(ficgp);
5598: fflush(fichtm);
5599: } /* end varevsij */
1.126 brouard 5600:
5601: /************ Variance of prevlim ******************/
1.235 brouard 5602: 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 5603: {
1.205 brouard 5604: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5605: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5606:
1.126 brouard 5607: double **dnewm,**doldm;
5608: int i, j, nhstepm, hstepm;
5609: double *xp;
5610: double *gp, *gm;
5611: double **gradg, **trgradg;
1.208 brouard 5612: double **mgm, **mgp;
1.126 brouard 5613: double age,agelim;
5614: int theta;
5615:
5616: pstamp(ficresvpl);
5617: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 ! brouard 5618: fprintf(ficresvpl,"# Age ");
! 5619: if(nresult >=1)
! 5620: fprintf(ficresvpl," Result# ");
1.126 brouard 5621: for(i=1; i<=nlstate;i++)
5622: fprintf(ficresvpl," %1d-%1d",i,i);
5623: fprintf(ficresvpl,"\n");
5624:
5625: xp=vector(1,npar);
5626: dnewm=matrix(1,nlstate,1,npar);
5627: doldm=matrix(1,nlstate,1,nlstate);
5628:
5629: hstepm=1*YEARM; /* Every year of age */
5630: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5631: agelim = AGESUP;
5632: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5633: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5634: if (stepm >= YEARM) hstepm=1;
5635: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5636: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5637: mgp=matrix(1,npar,1,nlstate);
5638: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5639: gp=vector(1,nlstate);
5640: gm=vector(1,nlstate);
5641:
5642: for(theta=1; theta <=npar; theta++){
5643: for(i=1; i<=npar; i++){ /* Computes gradient */
5644: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5645: }
1.209 brouard 5646: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5647: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5648: else
1.235 brouard 5649: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5650: for(i=1;i<=nlstate;i++){
1.126 brouard 5651: gp[i] = prlim[i][i];
1.208 brouard 5652: mgp[theta][i] = prlim[i][i];
5653: }
1.126 brouard 5654: for(i=1; i<=npar; i++) /* Computes gradient */
5655: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5656: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5657: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5658: else
1.235 brouard 5659: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5660: for(i=1;i<=nlstate;i++){
1.126 brouard 5661: gm[i] = prlim[i][i];
1.208 brouard 5662: mgm[theta][i] = prlim[i][i];
5663: }
1.126 brouard 5664: for(i=1;i<=nlstate;i++)
5665: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5666: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5667: } /* End theta */
5668:
5669: trgradg =matrix(1,nlstate,1,npar);
5670:
5671: for(j=1; j<=nlstate;j++)
5672: for(theta=1; theta <=npar; theta++)
5673: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5674: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5675: /* printf("\nmgm mgp %d ",(int)age); */
5676: /* for(j=1; j<=nlstate;j++){ */
5677: /* printf(" %d ",j); */
5678: /* for(theta=1; theta <=npar; theta++) */
5679: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5680: /* printf("\n "); */
5681: /* } */
5682: /* } */
5683: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5684: /* printf("\n gradg %d ",(int)age); */
5685: /* for(j=1; j<=nlstate;j++){ */
5686: /* printf("%d ",j); */
5687: /* for(theta=1; theta <=npar; theta++) */
5688: /* printf("%d %lf ",theta,gradg[theta][j]); */
5689: /* printf("\n "); */
5690: /* } */
5691: /* } */
1.126 brouard 5692:
5693: for(i=1;i<=nlstate;i++)
5694: varpl[i][(int)age] =0.;
1.209 brouard 5695: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 5696: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5697: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
5698: }else{
1.126 brouard 5699: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5700: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 5701: }
1.126 brouard 5702: for(i=1;i<=nlstate;i++)
5703: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
5704:
5705: fprintf(ficresvpl,"%.0f ",age );
1.241 ! brouard 5706: if(nresult >=1)
! 5707: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 5708: for(i=1; i<=nlstate;i++)
5709: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
5710: fprintf(ficresvpl,"\n");
5711: free_vector(gp,1,nlstate);
5712: free_vector(gm,1,nlstate);
1.208 brouard 5713: free_matrix(mgm,1,npar,1,nlstate);
5714: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 5715: free_matrix(gradg,1,npar,1,nlstate);
5716: free_matrix(trgradg,1,nlstate,1,npar);
5717: } /* End age */
5718:
5719: free_vector(xp,1,npar);
5720: free_matrix(doldm,1,nlstate,1,npar);
5721: free_matrix(dnewm,1,nlstate,1,nlstate);
5722:
5723: }
5724:
5725: /************ Variance of one-step probabilities ******************/
5726: 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 5727: {
5728: int i, j=0, k1, l1, tj;
5729: int k2, l2, j1, z1;
5730: int k=0, l;
5731: int first=1, first1, first2;
5732: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
5733: double **dnewm,**doldm;
5734: double *xp;
5735: double *gp, *gm;
5736: double **gradg, **trgradg;
5737: double **mu;
5738: double age, cov[NCOVMAX+1];
5739: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
5740: int theta;
5741: char fileresprob[FILENAMELENGTH];
5742: char fileresprobcov[FILENAMELENGTH];
5743: char fileresprobcor[FILENAMELENGTH];
5744: double ***varpij;
5745:
5746: strcpy(fileresprob,"PROB_");
5747: strcat(fileresprob,fileres);
5748: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
5749: printf("Problem with resultfile: %s\n", fileresprob);
5750: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
5751: }
5752: strcpy(fileresprobcov,"PROBCOV_");
5753: strcat(fileresprobcov,fileresu);
5754: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
5755: printf("Problem with resultfile: %s\n", fileresprobcov);
5756: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
5757: }
5758: strcpy(fileresprobcor,"PROBCOR_");
5759: strcat(fileresprobcor,fileresu);
5760: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
5761: printf("Problem with resultfile: %s\n", fileresprobcor);
5762: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
5763: }
5764: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5765: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5766: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5767: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5768: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5769: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5770: pstamp(ficresprob);
5771: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
5772: fprintf(ficresprob,"# Age");
5773: pstamp(ficresprobcov);
5774: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
5775: fprintf(ficresprobcov,"# Age");
5776: pstamp(ficresprobcor);
5777: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
5778: fprintf(ficresprobcor,"# Age");
1.126 brouard 5779:
5780:
1.222 brouard 5781: for(i=1; i<=nlstate;i++)
5782: for(j=1; j<=(nlstate+ndeath);j++){
5783: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
5784: fprintf(ficresprobcov," p%1d-%1d ",i,j);
5785: fprintf(ficresprobcor," p%1d-%1d ",i,j);
5786: }
5787: /* fprintf(ficresprob,"\n");
5788: fprintf(ficresprobcov,"\n");
5789: fprintf(ficresprobcor,"\n");
5790: */
5791: xp=vector(1,npar);
5792: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5793: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
5794: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
5795: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
5796: first=1;
5797: fprintf(ficgp,"\n# Routine varprob");
5798: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
5799: fprintf(fichtm,"\n");
5800:
5801: 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);
5802: 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);
5803: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 5804: and drawn. It helps understanding how is the covariance between two incidences.\
5805: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 5806: 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 5807: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
5808: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
5809: standard deviations wide on each axis. <br>\
5810: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
5811: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
5812: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
5813:
1.222 brouard 5814: cov[1]=1;
5815: /* tj=cptcoveff; */
1.225 brouard 5816: tj = (int) pow(2,cptcoveff);
1.222 brouard 5817: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
5818: j1=0;
1.224 brouard 5819: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 5820: if (cptcovn>0) {
5821: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 5822: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5823: fprintf(ficresprob, "**********\n#\n");
5824: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 5825: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5826: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 5827:
1.222 brouard 5828: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 5829: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5830: fprintf(ficgp, "**********\n#\n");
1.220 brouard 5831:
5832:
1.222 brouard 5833: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 5834: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5835: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 5836:
1.222 brouard 5837: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 5838: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5839: fprintf(ficresprobcor, "**********\n#");
5840: if(invalidvarcomb[j1]){
5841: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
5842: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
5843: continue;
5844: }
5845: }
5846: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
5847: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5848: gp=vector(1,(nlstate)*(nlstate+ndeath));
5849: gm=vector(1,(nlstate)*(nlstate+ndeath));
5850: for (age=bage; age<=fage; age ++){
5851: cov[2]=age;
5852: if(nagesqr==1)
5853: cov[3]= age*age;
5854: for (k=1; k<=cptcovn;k++) {
5855: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
5856: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
5857: * 1 1 1 1 1
5858: * 2 2 1 1 1
5859: * 3 1 2 1 1
5860: */
5861: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
5862: }
5863: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
5864: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
5865: for (k=1; k<=cptcovprod;k++)
5866: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 5867:
5868:
1.222 brouard 5869: for(theta=1; theta <=npar; theta++){
5870: for(i=1; i<=npar; i++)
5871: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 5872:
1.222 brouard 5873: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 5874:
1.222 brouard 5875: k=0;
5876: for(i=1; i<= (nlstate); i++){
5877: for(j=1; j<=(nlstate+ndeath);j++){
5878: k=k+1;
5879: gp[k]=pmmij[i][j];
5880: }
5881: }
1.220 brouard 5882:
1.222 brouard 5883: for(i=1; i<=npar; i++)
5884: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 5885:
1.222 brouard 5886: pmij(pmmij,cov,ncovmodel,xp,nlstate);
5887: k=0;
5888: for(i=1; i<=(nlstate); i++){
5889: for(j=1; j<=(nlstate+ndeath);j++){
5890: k=k+1;
5891: gm[k]=pmmij[i][j];
5892: }
5893: }
1.220 brouard 5894:
1.222 brouard 5895: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
5896: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
5897: }
1.126 brouard 5898:
1.222 brouard 5899: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
5900: for(theta=1; theta <=npar; theta++)
5901: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 5902:
1.222 brouard 5903: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
5904: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 5905:
1.222 brouard 5906: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 5907:
1.222 brouard 5908: k=0;
5909: for(i=1; i<=(nlstate); i++){
5910: for(j=1; j<=(nlstate+ndeath);j++){
5911: k=k+1;
5912: mu[k][(int) age]=pmmij[i][j];
5913: }
5914: }
5915: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
5916: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
5917: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 5918:
1.222 brouard 5919: /*printf("\n%d ",(int)age);
5920: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
5921: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
5922: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
5923: }*/
1.220 brouard 5924:
1.222 brouard 5925: fprintf(ficresprob,"\n%d ",(int)age);
5926: fprintf(ficresprobcov,"\n%d ",(int)age);
5927: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 5928:
1.222 brouard 5929: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
5930: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
5931: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
5932: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
5933: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
5934: }
5935: i=0;
5936: for (k=1; k<=(nlstate);k++){
5937: for (l=1; l<=(nlstate+ndeath);l++){
5938: i++;
5939: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
5940: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
5941: for (j=1; j<=i;j++){
5942: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
5943: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
5944: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
5945: }
5946: }
5947: }/* end of loop for state */
5948: } /* end of loop for age */
5949: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
5950: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
5951: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
5952: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
5953:
5954: /* Confidence intervalle of pij */
5955: /*
5956: fprintf(ficgp,"\nunset parametric;unset label");
5957: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
5958: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
5959: 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);
5960: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
5961: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
5962: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
5963: */
5964:
5965: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
5966: first1=1;first2=2;
5967: for (k2=1; k2<=(nlstate);k2++){
5968: for (l2=1; l2<=(nlstate+ndeath);l2++){
5969: if(l2==k2) continue;
5970: j=(k2-1)*(nlstate+ndeath)+l2;
5971: for (k1=1; k1<=(nlstate);k1++){
5972: for (l1=1; l1<=(nlstate+ndeath);l1++){
5973: if(l1==k1) continue;
5974: i=(k1-1)*(nlstate+ndeath)+l1;
5975: if(i<=j) continue;
5976: for (age=bage; age<=fage; age ++){
5977: if ((int)age %5==0){
5978: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
5979: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
5980: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
5981: mu1=mu[i][(int) age]/stepm*YEARM ;
5982: mu2=mu[j][(int) age]/stepm*YEARM;
5983: c12=cv12/sqrt(v1*v2);
5984: /* Computing eigen value of matrix of covariance */
5985: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
5986: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
5987: if ((lc2 <0) || (lc1 <0) ){
5988: if(first2==1){
5989: first1=0;
5990: 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);
5991: }
5992: 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);
5993: /* lc1=fabs(lc1); */ /* If we want to have them positive */
5994: /* lc2=fabs(lc2); */
5995: }
1.220 brouard 5996:
1.222 brouard 5997: /* Eigen vectors */
5998: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
5999: /*v21=sqrt(1.-v11*v11); *//* error */
6000: v21=(lc1-v1)/cv12*v11;
6001: v12=-v21;
6002: v22=v11;
6003: tnalp=v21/v11;
6004: if(first1==1){
6005: first1=0;
6006: 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);
6007: }
6008: 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);
6009: /*printf(fignu*/
6010: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6011: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6012: if(first==1){
6013: first=0;
6014: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6015: fprintf(ficgp,"\nset parametric;unset label");
6016: 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);
6017: fprintf(ficgp,"\nset ter svg size 640, 480");
6018: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6019: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6020: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6021: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6022: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6023: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6024: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6025: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6026: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6027: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6028: 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", \
6029: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6030: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6031: }else{
6032: first=0;
6033: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6034: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6035: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6036: 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", \
6037: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6038: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6039: }/* if first */
6040: } /* age mod 5 */
6041: } /* end loop age */
6042: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6043: first=1;
6044: } /*l12 */
6045: } /* k12 */
6046: } /*l1 */
6047: }/* k1 */
6048: } /* loop on combination of covariates j1 */
6049: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6050: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6051: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6052: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6053: free_vector(xp,1,npar);
6054: fclose(ficresprob);
6055: fclose(ficresprobcov);
6056: fclose(ficresprobcor);
6057: fflush(ficgp);
6058: fflush(fichtmcov);
6059: }
1.126 brouard 6060:
6061:
6062: /******************* Printing html file ***********/
1.201 brouard 6063: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6064: int lastpass, int stepm, int weightopt, char model[],\
6065: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6066: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6067: double jprev1, double mprev1,double anprev1, double dateprev1, \
6068: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6069: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6070:
6071: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6072: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6073: </ul>");
1.237 brouard 6074: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6075: </ul>", model);
1.214 brouard 6076: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6077: 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",
6078: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6079: 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 6080: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6081: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6082: fprintf(fichtm,"\
6083: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6084: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6085: fprintf(fichtm,"\
1.217 brouard 6086: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6087: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6088: fprintf(fichtm,"\
1.126 brouard 6089: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6090: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6091: fprintf(fichtm,"\
1.217 brouard 6092: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6093: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6094: fprintf(fichtm,"\
1.211 brouard 6095: - (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 6096: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6097: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6098: if(prevfcast==1){
6099: fprintf(fichtm,"\
6100: - Prevalence projections by age and states: \
1.201 brouard 6101: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6102: }
1.126 brouard 6103:
1.222 brouard 6104: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6105:
1.225 brouard 6106: m=pow(2,cptcoveff);
1.222 brouard 6107: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6108:
1.222 brouard 6109: jj1=0;
1.237 brouard 6110:
6111: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 ! brouard 6112: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.237 brouard 6113: if(TKresult[nres]!= k1)
6114: continue;
1.220 brouard 6115:
1.222 brouard 6116: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6117: jj1++;
6118: if (cptcovn > 0) {
6119: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6120: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6121: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6122: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6123: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6124: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6125: }
1.237 brouard 6126: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6127: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6128: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6129: }
6130:
1.230 brouard 6131: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6132: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6133: if(invalidvarcomb[k1]){
6134: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6135: printf("\nCombination (%d) ignored because no cases \n",k1);
6136: continue;
6137: }
6138: }
6139: /* aij, bij */
1.241 ! brouard 6140: 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> \
! 6141: <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 6142: /* Pij */
1.241 ! brouard 6143: 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> \
! 6144: <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 6145: /* Quasi-incidences */
6146: 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 6147: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6148: 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 6149: 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> \
! 6150: <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 6151: /* Survival functions (period) in state j */
6152: for(cpt=1; cpt<=nlstate;cpt++){
1.241 ! brouard 6153: 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> \
! 6154: <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 6155: }
6156: /* State specific survival functions (period) */
6157: for(cpt=1; cpt<=nlstate;cpt++){
6158: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6159: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 ! brouard 6160: <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 6161: }
6162: /* Period (stable) prevalence in each health state */
6163: for(cpt=1; cpt<=nlstate;cpt++){
1.241 ! brouard 6164: 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> \
! 6165: <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 6166: }
6167: if(backcast==1){
6168: /* Period (stable) back prevalence in each health state */
6169: for(cpt=1; cpt<=nlstate;cpt++){
1.241 ! brouard 6170: 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> \
! 6171: <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 6172: }
1.217 brouard 6173: }
1.222 brouard 6174: if(prevfcast==1){
6175: /* Projection of prevalence up to period (stable) prevalence in each health state */
6176: for(cpt=1; cpt<=nlstate;cpt++){
1.241 ! brouard 6177: 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> \
! 6178: <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 6179: }
6180: }
1.220 brouard 6181:
1.222 brouard 6182: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 ! brouard 6183: 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> \
! 6184: <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 6185: }
6186: /* } /\* end i1 *\/ */
6187: }/* End k1 */
6188: fprintf(fichtm,"</ul>");
1.126 brouard 6189:
1.222 brouard 6190: fprintf(fichtm,"\
1.126 brouard 6191: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6192: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6193: - 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 6194: But because parameters are usually highly correlated (a higher incidence of disability \
6195: and a higher incidence of recovery can give very close observed transition) it might \
6196: be very useful to look not only at linear confidence intervals estimated from the \
6197: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6198: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6199: covariance matrix of the one-step probabilities. \
6200: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6201:
1.222 brouard 6202: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6203: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6204: fprintf(fichtm,"\
1.126 brouard 6205: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6206: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6207:
1.222 brouard 6208: fprintf(fichtm,"\
1.126 brouard 6209: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6210: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6211: fprintf(fichtm,"\
1.126 brouard 6212: - 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): \
6213: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6214: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6215: fprintf(fichtm,"\
1.126 brouard 6216: - (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): \
6217: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6218: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6219: fprintf(fichtm,"\
1.128 brouard 6220: - 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 6221: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6222: fprintf(fichtm,"\
1.128 brouard 6223: - 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 6224: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6225: fprintf(fichtm,"\
1.126 brouard 6226: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6227: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6228:
6229: /* if(popforecast==1) fprintf(fichtm,"\n */
6230: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6231: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6232: /* <br>",fileres,fileres,fileres,fileres); */
6233: /* else */
6234: /* 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 6235: fflush(fichtm);
6236: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6237:
1.225 brouard 6238: m=pow(2,cptcoveff);
1.222 brouard 6239: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6240:
1.222 brouard 6241: jj1=0;
1.237 brouard 6242:
1.241 ! brouard 6243: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6244: for(k1=1; k1<=m;k1++){
1.237 brouard 6245: if(TKresult[nres]!= k1)
6246: continue;
1.222 brouard 6247: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6248: jj1++;
1.126 brouard 6249: if (cptcovn > 0) {
6250: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6251: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6252: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6253: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6254: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6255: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6256: }
6257:
1.126 brouard 6258: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6259:
1.222 brouard 6260: if(invalidvarcomb[k1]){
6261: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6262: continue;
6263: }
1.126 brouard 6264: }
6265: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6266: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 ! brouard 6267: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
! 6268: <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 6269: }
6270: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6271: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6272: true period expectancies (those weighted with period prevalences are also\
6273: drawn in addition to the population based expectancies computed using\
1.241 ! brouard 6274: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
! 6275: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6276: /* } /\* end i1 *\/ */
6277: }/* End k1 */
1.241 ! brouard 6278: }/* End nres */
1.222 brouard 6279: fprintf(fichtm,"</ul>");
6280: fflush(fichtm);
1.126 brouard 6281: }
6282:
6283: /******************* Gnuplot file **************/
1.223 brouard 6284: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6285:
6286: char dirfileres[132],optfileres[132];
1.223 brouard 6287: char gplotcondition[132];
1.237 brouard 6288: 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 6289: int lv=0, vlv=0, kl=0;
1.130 brouard 6290: int ng=0;
1.201 brouard 6291: int vpopbased;
1.223 brouard 6292: int ioffset; /* variable offset for columns */
1.235 brouard 6293: int nres=0; /* Index of resultline */
1.219 brouard 6294:
1.126 brouard 6295: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6296: /* printf("Problem with file %s",optionfilegnuplot); */
6297: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6298: /* } */
6299:
6300: /*#ifdef windows */
6301: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6302: /*#endif */
1.225 brouard 6303: m=pow(2,cptcoveff);
1.126 brouard 6304:
1.202 brouard 6305: /* Contribution to likelihood */
6306: /* Plot the probability implied in the likelihood */
1.223 brouard 6307: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6308: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6309: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6310: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6311: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6312: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6313: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6314: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6315: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6316: 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));
6317: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6318: 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));
6319: for (i=1; i<= nlstate ; i ++) {
6320: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6321: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6322: 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);
6323: for (j=2; j<= nlstate+ndeath ; j ++) {
6324: 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);
6325: }
6326: fprintf(ficgp,";\nset out; unset ylabel;\n");
6327: }
6328: /* 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 */
6329: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6330: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6331: fprintf(ficgp,"\nset out;unset log\n");
6332: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6333:
1.126 brouard 6334: strcpy(dirfileres,optionfilefiname);
6335: strcpy(optfileres,"vpl");
1.223 brouard 6336: /* 1eme*/
1.238 brouard 6337: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6338: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6339: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6340: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
6341: if(TKresult[nres]!= k1)
6342: continue;
6343: /* We are interested in selected combination by the resultline */
6344: printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6345: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6346: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6347: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6348: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6349: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6350: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6351: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6352: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
6353: printf(" V%d=%d ",Tvaraff[k],vlv);
6354: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6355: }
6356: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6357: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6358: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6359: }
6360: printf("\n#\n");
6361: fprintf(ficgp,"\n#\n");
6362: if(invalidvarcomb[k1]){
6363: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6364: continue;
6365: }
1.235 brouard 6366:
1.241 ! brouard 6367: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
! 6368: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
! 6369: 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 6370:
1.238 brouard 6371: for (i=1; i<= nlstate ; i ++) {
6372: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6373: else fprintf(ficgp," %%*lf (%%*lf)");
6374: }
1.241 ! brouard 6375: fprintf(ficgp,"\" t\"Period (stable) prevalence\" w l lt 0,\"%s\" every :::%d::%d u 1:($2==%d ? $4+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6376: for (i=1; i<= nlstate ; i ++) {
6377: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6378: else fprintf(ficgp," %%*lf (%%*lf)");
6379: }
1.241 ! brouard 6380: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $4-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6381: for (i=1; i<= nlstate ; i ++) {
6382: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6383: else fprintf(ficgp," %%*lf (%%*lf)");
6384: }
6385: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6386: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6387: /* 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); */
1.241 ! brouard 6388: fprintf(ficgp,",\"%s\" u ($2==%d ?$1:1/0):(",subdirf2(fileresu,"PLB_"),nres); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6389: if(cptcoveff ==0){
6390: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line ", 2+(cpt-1), cpt );
6391: }else{
6392: kl=0;
6393: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6394: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6395: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6396: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6397: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6398: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6399: kl++;
1.238 brouard 6400: /* 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 *\/ */
6401: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6402: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6403: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6404: if(k==cptcoveff){
6405: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
6406: 4+(cpt-1), cpt ); /* 4 or 6 ?*/
6407: }else{
6408: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6409: kl++;
6410: }
6411: } /* end covariate */
6412: } /* end if no covariate */
6413: } /* end if backcast */
6414: fprintf(ficgp,"\nset out \n");
6415: } /* nres */
1.201 brouard 6416: } /* k1 */
6417: } /* cpt */
1.235 brouard 6418:
6419:
1.126 brouard 6420: /*2 eme*/
1.238 brouard 6421: for (k1=1; k1<= m ; k1 ++){
6422: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6423: if(TKresult[nres]!= k1)
6424: continue;
6425: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6426: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6427: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6428: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6429: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6430: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6431: vlv= nbcode[Tvaraff[k]][lv];
6432: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6433: }
1.237 brouard 6434: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6435: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6436: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6437: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6438: }
1.211 brouard 6439: fprintf(ficgp,"\n#\n");
1.223 brouard 6440: if(invalidvarcomb[k1]){
6441: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6442: continue;
6443: }
1.219 brouard 6444:
1.241 ! brouard 6445: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6446: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6447: if(vpopbased==0)
6448: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6449: else
6450: fprintf(ficgp,"\nreplot ");
6451: for (i=1; i<= nlstate+1 ; i ++) {
6452: k=2*i;
6453: 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);
6454: for (j=1; j<= nlstate+1 ; j ++) {
6455: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6456: else fprintf(ficgp," %%*lf (%%*lf)");
6457: }
6458: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6459: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6460: 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);
6461: for (j=1; j<= nlstate+1 ; j ++) {
6462: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6463: else fprintf(ficgp," %%*lf (%%*lf)");
6464: }
6465: fprintf(ficgp,"\" t\"\" w l lt 0,");
6466: 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);
6467: for (j=1; j<= nlstate+1 ; j ++) {
6468: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6469: else fprintf(ficgp," %%*lf (%%*lf)");
6470: }
6471: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6472: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6473: } /* state */
6474: } /* vpopbased */
6475: fprintf(ficgp,"\nset out;set out \"%s_%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1); /* Buggy gnuplot */
6476: } /* end nres */
6477: } /* k1 end 2 eme*/
6478:
6479:
6480: /*3eme*/
6481: for (k1=1; k1<= m ; k1 ++){
6482: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.240 brouard 6483: if(TKresult[nres]!= k1)
1.238 brouard 6484: continue;
6485:
6486: for (cpt=1; cpt<= nlstate ; cpt ++) {
6487: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6488: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6489: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6490: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6491: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6492: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6493: vlv= nbcode[Tvaraff[k]][lv];
6494: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6495: }
6496: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6497: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6498: }
6499: fprintf(ficgp,"\n#\n");
6500: if(invalidvarcomb[k1]){
6501: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6502: continue;
6503: }
6504:
6505: /* k=2+nlstate*(2*cpt-2); */
6506: k=2+(nlstate+1)*(cpt-1);
1.241 ! brouard 6507: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6508: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6509: 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 6510: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6511: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6512: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6513: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6514: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6515: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6516:
1.238 brouard 6517: */
6518: for (i=1; i< nlstate ; i ++) {
6519: 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);
6520: /* 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 6521:
1.238 brouard 6522: }
6523: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6524: }
6525: } /* end nres */
6526: } /* end kl 3eme */
1.126 brouard 6527:
1.223 brouard 6528: /* 4eme */
1.201 brouard 6529: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6530: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6531: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6532: if(TKresult[nres]!= k1)
1.223 brouard 6533: continue;
1.238 brouard 6534: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6535: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6536: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6537: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6538: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6539: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6540: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6541: vlv= nbcode[Tvaraff[k]][lv];
6542: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6543: }
6544: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6545: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6546: }
6547: fprintf(ficgp,"\n#\n");
6548: if(invalidvarcomb[k1]){
6549: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6550: continue;
1.223 brouard 6551: }
1.238 brouard 6552:
1.241 ! brouard 6553: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6554: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6555: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6556: k=3;
6557: for (i=1; i<= nlstate ; i ++){
6558: if(i==1){
6559: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6560: }else{
6561: fprintf(ficgp,", '' ");
6562: }
6563: l=(nlstate+ndeath)*(i-1)+1;
6564: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6565: for (j=2; j<= nlstate+ndeath ; j ++)
6566: fprintf(ficgp,"+$%d",k+l+j-1);
6567: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6568: } /* nlstate */
6569: fprintf(ficgp,"\nset out\n");
6570: } /* end cpt state*/
6571: } /* end nres */
6572: } /* end covariate k1 */
6573:
1.220 brouard 6574: /* 5eme */
1.201 brouard 6575: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6576: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6577: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6578: if(TKresult[nres]!= k1)
1.227 brouard 6579: continue;
1.238 brouard 6580: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6581: 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);
6582: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6583: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6584: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6585: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6586: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6587: vlv= nbcode[Tvaraff[k]][lv];
6588: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6589: }
6590: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6591: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6592: }
6593: fprintf(ficgp,"\n#\n");
6594: if(invalidvarcomb[k1]){
6595: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6596: continue;
6597: }
1.227 brouard 6598:
1.241 ! brouard 6599: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6600: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6601: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6602: k=3;
6603: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6604: if(j==1)
6605: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6606: else
6607: fprintf(ficgp,", '' ");
6608: l=(nlstate+ndeath)*(cpt-1) +j;
6609: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6610: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6611: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6612: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6613: } /* nlstate */
6614: fprintf(ficgp,", '' ");
6615: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6616: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6617: l=(nlstate+ndeath)*(cpt-1) +j;
6618: if(j < nlstate)
6619: fprintf(ficgp,"$%d +",k+l);
6620: else
6621: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6622: }
6623: fprintf(ficgp,"\nset out\n");
6624: } /* end cpt state*/
6625: } /* end covariate */
6626: } /* end nres */
1.227 brouard 6627:
1.220 brouard 6628: /* 6eme */
1.202 brouard 6629: /* CV preval stable (period) for each covariate */
1.237 brouard 6630: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6631: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6632: if(TKresult[nres]!= k1)
6633: continue;
1.153 brouard 6634: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6635:
1.211 brouard 6636: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6637: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6638: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6639: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6640: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6641: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6642: vlv= nbcode[Tvaraff[k]][lv];
6643: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6644: }
1.237 brouard 6645: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6646: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6647: }
1.211 brouard 6648: fprintf(ficgp,"\n#\n");
1.223 brouard 6649: if(invalidvarcomb[k1]){
1.227 brouard 6650: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6651: continue;
1.223 brouard 6652: }
1.227 brouard 6653:
1.241 ! brouard 6654: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6655: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6656: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6657: k=3; /* Offset */
1.153 brouard 6658: for (i=1; i<= nlstate ; i ++){
1.227 brouard 6659: if(i==1)
6660: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6661: else
6662: fprintf(ficgp,", '' ");
6663: l=(nlstate+ndeath)*(i-1)+1;
6664: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6665: for (j=2; j<= nlstate ; j ++)
6666: fprintf(ficgp,"+$%d",k+l+j-1);
6667: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6668: } /* nlstate */
1.201 brouard 6669: fprintf(ficgp,"\nset out\n");
1.153 brouard 6670: } /* end cpt state*/
6671: } /* end covariate */
1.227 brouard 6672:
6673:
1.220 brouard 6674: /* 7eme */
1.218 brouard 6675: if(backcast == 1){
1.217 brouard 6676: /* CV back preval stable (period) for each covariate */
1.237 brouard 6677: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6678: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6679: if(TKresult[nres]!= k1)
6680: continue;
1.218 brouard 6681: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6682: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
6683: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6684: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6685: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6686: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6687: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6688: vlv= nbcode[Tvaraff[k]][lv];
6689: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6690: }
1.237 brouard 6691: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6692: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6693: }
1.227 brouard 6694: fprintf(ficgp,"\n#\n");
6695: if(invalidvarcomb[k1]){
6696: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6697: continue;
6698: }
6699:
1.241 ! brouard 6700: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 6701: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6702: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6703: k=3; /* Offset */
6704: for (i=1; i<= nlstate ; i ++){
6705: if(i==1)
6706: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
6707: else
6708: fprintf(ficgp,", '' ");
6709: /* l=(nlstate+ndeath)*(i-1)+1; */
6710: l=(nlstate+ndeath)*(cpt-1)+1;
6711: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
6712: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
6713: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+(cpt-1)+i-1); /* a vérifier */
6714: /* for (j=2; j<= nlstate ; j ++) */
6715: /* fprintf(ficgp,"+$%d",k+l+j-1); */
6716: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
6717: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
6718: } /* nlstate */
6719: fprintf(ficgp,"\nset out\n");
1.218 brouard 6720: } /* end cpt state*/
6721: } /* end covariate */
6722: } /* End if backcast */
6723:
1.223 brouard 6724: /* 8eme */
1.218 brouard 6725: if(prevfcast==1){
6726: /* Projection from cross-sectional to stable (period) for each covariate */
6727:
1.237 brouard 6728: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6729: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6730: if(TKresult[nres]!= k1)
6731: continue;
1.211 brouard 6732: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6733: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
6734: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
6735: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6736: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6737: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6738: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6739: vlv= nbcode[Tvaraff[k]][lv];
6740: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6741: }
1.237 brouard 6742: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6743: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6744: }
1.227 brouard 6745: fprintf(ficgp,"\n#\n");
6746: if(invalidvarcomb[k1]){
6747: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6748: continue;
6749: }
6750:
6751: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 ! brouard 6752: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 6753: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 6754: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6755: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
6756: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6757: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6758: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6759: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6760: if(i==1){
6761: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
6762: }else{
6763: fprintf(ficgp,",\\\n '' ");
6764: }
6765: if(cptcoveff ==0){ /* No covariate */
6766: ioffset=2; /* Age is in 2 */
6767: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6768: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6769: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6770: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6771: fprintf(ficgp," u %d:(", ioffset);
6772: if(i==nlstate+1)
6773: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
6774: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6775: else
6776: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
6777: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6778: }else{ /* more than 2 covariates */
6779: if(cptcoveff ==1){
6780: ioffset=4; /* Age is in 4 */
6781: }else{
6782: ioffset=6; /* Age is in 6 */
6783: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6784: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6785: }
6786: fprintf(ficgp," u %d:(",ioffset);
6787: kl=0;
6788: strcpy(gplotcondition,"(");
6789: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
6790: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
6791: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6792: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6793: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6794: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
6795: kl++;
6796: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
6797: kl++;
6798: if(k <cptcoveff && cptcoveff>1)
6799: sprintf(gplotcondition+strlen(gplotcondition)," && ");
6800: }
6801: strcpy(gplotcondition+strlen(gplotcondition),")");
6802: /* 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 *\/ */
6803: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6804: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6805: /* '' 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*/
6806: if(i==nlstate+1){
6807: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
6808: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6809: }else{
6810: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
6811: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6812: }
6813: } /* end if covariate */
6814: } /* nlstate */
6815: fprintf(ficgp,"\nset out\n");
1.223 brouard 6816: } /* end cpt state*/
6817: } /* end covariate */
6818: } /* End if prevfcast */
1.227 brouard 6819:
6820:
1.238 brouard 6821: /* 9eme writing MLE parameters */
6822: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 6823: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 6824: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 6825: for(k=1; k <=(nlstate+ndeath); k++){
6826: if (k != i) {
1.227 brouard 6827: fprintf(ficgp,"# current state %d\n",k);
6828: for(j=1; j <=ncovmodel; j++){
6829: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
6830: jk++;
6831: }
6832: fprintf(ficgp,"\n");
1.126 brouard 6833: }
6834: }
1.223 brouard 6835: }
1.187 brouard 6836: fprintf(ficgp,"##############\n#\n");
1.227 brouard 6837:
1.145 brouard 6838: /*goto avoid;*/
1.238 brouard 6839: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
6840: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 6841: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
6842: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
6843: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
6844: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
6845: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6846: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6847: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6848: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6849: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
6850: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6851: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
6852: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
6853: fprintf(ficgp,"#\n");
1.223 brouard 6854: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 6855: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 6856: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 6857: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 6858: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
6859: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
6860: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6861: if(TKresult[nres]!= jk)
6862: continue;
6863: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
6864: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6865: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6866: }
6867: fprintf(ficgp,"\n#\n");
1.241 ! brouard 6868: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 6869: fprintf(ficgp,"\nset ter svg size 640, 480 ");
6870: if (ng==1){
6871: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
6872: fprintf(ficgp,"\nunset log y");
6873: }else if (ng==2){
6874: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
6875: fprintf(ficgp,"\nset log y");
6876: }else if (ng==3){
6877: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
6878: fprintf(ficgp,"\nset log y");
6879: }else
6880: fprintf(ficgp,"\nunset title ");
6881: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
6882: i=1;
6883: for(k2=1; k2<=nlstate; k2++) {
6884: k3=i;
6885: for(k=1; k<=(nlstate+ndeath); k++) {
6886: if (k != k2){
6887: switch( ng) {
6888: case 1:
6889: if(nagesqr==0)
6890: fprintf(ficgp," p%d+p%d*x",i,i+1);
6891: else /* nagesqr =1 */
6892: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
6893: break;
6894: case 2: /* ng=2 */
6895: if(nagesqr==0)
6896: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
6897: else /* nagesqr =1 */
6898: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
6899: break;
6900: case 3:
6901: if(nagesqr==0)
6902: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
6903: else /* nagesqr =1 */
6904: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
6905: break;
6906: }
6907: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 6908: ijp=1; /* product no age */
6909: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
6910: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 6911: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 6912: if(j==Tage[ij]) { /* Product by age */
6913: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 6914: if(DummyV[j]==0){
1.237 brouard 6915: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
6916: }else{ /* quantitative */
6917: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
6918: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
6919: }
6920: ij++;
6921: }
6922: }else if(j==Tprod[ijp]) { /* */
6923: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
6924: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 6925: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
6926: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 6927: /* 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)]); */
6928: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
6929: }else{ /* Vn is dummy and Vm is quanti */
6930: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
6931: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
6932: }
6933: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 6934: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 6935: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
6936: }else{ /* Both quanti */
6937: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
6938: }
6939: }
1.238 brouard 6940: ijp++;
1.237 brouard 6941: }
6942: } else{ /* simple covariate */
6943: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
6944: if(Dummy[j]==0){
6945: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
6946: }else{ /* quantitative */
6947: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 6948: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
6949: }
1.237 brouard 6950: } /* end simple */
6951: } /* end j */
1.223 brouard 6952: }else{
6953: i=i-ncovmodel;
6954: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
6955: fprintf(ficgp," (1.");
6956: }
1.227 brouard 6957:
1.223 brouard 6958: if(ng != 1){
6959: fprintf(ficgp,")/(1");
1.227 brouard 6960:
1.223 brouard 6961: for(k1=1; k1 <=nlstate; k1++){
6962: if(nagesqr==0)
6963: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
6964: else /* nagesqr =1 */
6965: 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 6966:
1.223 brouard 6967: ij=1;
6968: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 6969: if((j-2)==Tage[ij]) { /* Bug valgrind */
6970: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 6971: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
6972: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
6973: ij++;
6974: }
6975: }
6976: else
1.225 brouard 6977: 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 6978: }
6979: fprintf(ficgp,")");
6980: }
6981: fprintf(ficgp,")");
6982: if(ng ==2)
6983: fprintf(ficgp," t \"p%d%d\" ", k2,k);
6984: else /* ng= 3 */
6985: fprintf(ficgp," t \"i%d%d\" ", k2,k);
6986: }else{ /* end ng <> 1 */
6987: if( k !=k2) /* logit p11 is hard to draw */
6988: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
6989: }
6990: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
6991: fprintf(ficgp,",");
6992: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
6993: fprintf(ficgp,",");
6994: i=i+ncovmodel;
6995: } /* end k */
6996: } /* end k2 */
6997: fprintf(ficgp,"\n set out\n");
6998: } /* end jk */
6999: } /* end ng */
7000: /* avoid: */
7001: fflush(ficgp);
1.126 brouard 7002: } /* end gnuplot */
7003:
7004:
7005: /*************** Moving average **************/
1.219 brouard 7006: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7007: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7008:
1.222 brouard 7009: int i, cpt, cptcod;
7010: int modcovmax =1;
7011: int mobilavrange, mob;
7012: int iage=0;
7013:
7014: double sum=0.;
7015: double age;
7016: double *sumnewp, *sumnewm;
7017: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7018:
7019:
1.225 brouard 7020: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7021: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7022:
7023: sumnewp = vector(1,ncovcombmax);
7024: sumnewm = vector(1,ncovcombmax);
7025: agemingood = vector(1,ncovcombmax);
7026: agemaxgood = vector(1,ncovcombmax);
7027:
7028: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7029: sumnewm[cptcod]=0.;
7030: sumnewp[cptcod]=0.;
7031: agemingood[cptcod]=0;
7032: agemaxgood[cptcod]=0;
7033: }
7034: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7035:
7036: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7037: if(mobilav==1) mobilavrange=5; /* default */
7038: else mobilavrange=mobilav;
7039: for (age=bage; age<=fage; age++)
7040: for (i=1; i<=nlstate;i++)
7041: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7042: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7043: /* We keep the original values on the extreme ages bage, fage and for
7044: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7045: we use a 5 terms etc. until the borders are no more concerned.
7046: */
7047: for (mob=3;mob <=mobilavrange;mob=mob+2){
7048: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7049: for (i=1; i<=nlstate;i++){
7050: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7051: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7052: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7053: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7054: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7055: }
7056: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7057: }
7058: }
7059: }/* end age */
7060: }/* end mob */
7061: }else
7062: return -1;
7063: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7064: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7065: if(invalidvarcomb[cptcod]){
7066: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7067: continue;
7068: }
1.219 brouard 7069:
1.222 brouard 7070: agemingood[cptcod]=fage-(mob-1)/2;
7071: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7072: sumnewm[cptcod]=0.;
7073: for (i=1; i<=nlstate;i++){
7074: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7075: }
7076: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7077: agemingood[cptcod]=age;
7078: }else{ /* bad */
7079: for (i=1; i<=nlstate;i++){
7080: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7081: } /* i */
7082: } /* end bad */
7083: }/* age */
7084: sum=0.;
7085: for (i=1; i<=nlstate;i++){
7086: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7087: }
7088: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7089: 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);
7090: /* for (i=1; i<=nlstate;i++){ */
7091: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7092: /* } /\* i *\/ */
7093: } /* end bad */
7094: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7095: /* From youngest, finding the oldest wrong */
7096: agemaxgood[cptcod]=bage+(mob-1)/2;
7097: for (age=bage+(mob-1)/2; age<=fage; age++){
7098: sumnewm[cptcod]=0.;
7099: for (i=1; i<=nlstate;i++){
7100: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7101: }
7102: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7103: agemaxgood[cptcod]=age;
7104: }else{ /* bad */
7105: for (i=1; i<=nlstate;i++){
7106: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7107: } /* i */
7108: } /* end bad */
7109: }/* age */
7110: sum=0.;
7111: for (i=1; i<=nlstate;i++){
7112: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7113: }
7114: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7115: 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);
7116: /* for (i=1; i<=nlstate;i++){ */
7117: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7118: /* } /\* i *\/ */
7119: } /* end bad */
7120:
7121: for (age=bage; age<=fage; age++){
1.235 brouard 7122: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7123: sumnewp[cptcod]=0.;
7124: sumnewm[cptcod]=0.;
7125: for (i=1; i<=nlstate;i++){
7126: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7127: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7128: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7129: }
7130: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7131: }
7132: /* printf("\n"); */
7133: /* } */
7134: /* brutal averaging */
7135: for (i=1; i<=nlstate;i++){
7136: for (age=1; age<=bage; age++){
7137: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7138: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7139: }
7140: for (age=fage; age<=AGESUP; age++){
7141: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7142: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7143: }
7144: } /* end i status */
7145: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7146: for (age=1; age<=AGESUP; age++){
7147: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7148: mobaverage[(int)age][i][cptcod]=0.;
7149: }
7150: }
7151: }/* end cptcod */
7152: free_vector(sumnewm,1, ncovcombmax);
7153: free_vector(sumnewp,1, ncovcombmax);
7154: free_vector(agemaxgood,1, ncovcombmax);
7155: free_vector(agemingood,1, ncovcombmax);
7156: return 0;
7157: }/* End movingaverage */
1.218 brouard 7158:
1.126 brouard 7159:
7160: /************** Forecasting ******************/
1.235 brouard 7161: 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 7162: /* proj1, year, month, day of starting projection
7163: agemin, agemax range of age
7164: dateprev1 dateprev2 range of dates during which prevalence is computed
7165: anproj2 year of en of projection (same day and month as proj1).
7166: */
1.235 brouard 7167: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7168: double agec; /* generic age */
7169: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7170: double *popeffectif,*popcount;
7171: double ***p3mat;
1.218 brouard 7172: /* double ***mobaverage; */
1.126 brouard 7173: char fileresf[FILENAMELENGTH];
7174:
7175: agelim=AGESUP;
1.211 brouard 7176: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7177: in each health status at the date of interview (if between dateprev1 and dateprev2).
7178: We still use firstpass and lastpass as another selection.
7179: */
1.214 brouard 7180: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7181: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7182:
1.201 brouard 7183: strcpy(fileresf,"F_");
7184: strcat(fileresf,fileresu);
1.126 brouard 7185: if((ficresf=fopen(fileresf,"w"))==NULL) {
7186: printf("Problem with forecast resultfile: %s\n", fileresf);
7187: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7188: }
1.235 brouard 7189: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7190: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7191:
1.225 brouard 7192: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7193:
7194:
7195: stepsize=(int) (stepm+YEARM-1)/YEARM;
7196: if (stepm<=12) stepsize=1;
7197: if(estepm < stepm){
7198: printf ("Problem %d lower than %d\n",estepm, stepm);
7199: }
7200: else hstepm=estepm;
7201:
7202: hstepm=hstepm/stepm;
7203: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7204: fractional in yp1 */
7205: anprojmean=yp;
7206: yp2=modf((yp1*12),&yp);
7207: mprojmean=yp;
7208: yp1=modf((yp2*30.5),&yp);
7209: jprojmean=yp;
7210: if(jprojmean==0) jprojmean=1;
7211: if(mprojmean==0) jprojmean=1;
7212:
1.227 brouard 7213: i1=pow(2,cptcoveff);
1.126 brouard 7214: if (cptcovn < 1){i1=1;}
7215:
7216: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7217:
7218: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7219:
1.126 brouard 7220: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7221: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7222: for(k=1; k<=i1;k++){
7223: if(TKresult[nres]!= k)
7224: continue;
1.227 brouard 7225: if(invalidvarcomb[k]){
7226: printf("\nCombination (%d) projection ignored because no cases \n",k);
7227: continue;
7228: }
7229: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7230: for(j=1;j<=cptcoveff;j++) {
7231: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7232: }
1.235 brouard 7233: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7234: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7235: }
1.227 brouard 7236: fprintf(ficresf," yearproj age");
7237: for(j=1; j<=nlstate+ndeath;j++){
7238: for(i=1; i<=nlstate;i++)
7239: fprintf(ficresf," p%d%d",i,j);
7240: fprintf(ficresf," wp.%d",j);
7241: }
7242: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7243: fprintf(ficresf,"\n");
7244: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7245: for (agec=fage; agec>=(ageminpar-1); agec--){
7246: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7247: nhstepm = nhstepm/hstepm;
7248: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7249: oldm=oldms;savm=savms;
1.235 brouard 7250: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7251:
7252: for (h=0; h<=nhstepm; h++){
7253: if (h*hstepm/YEARM*stepm ==yearp) {
7254: fprintf(ficresf,"\n");
7255: for(j=1;j<=cptcoveff;j++)
7256: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7257: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7258: }
7259: for(j=1; j<=nlstate+ndeath;j++) {
7260: ppij=0.;
7261: for(i=1; i<=nlstate;i++) {
7262: if (mobilav==1)
7263: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7264: else {
7265: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7266: }
7267: if (h*hstepm/YEARM*stepm== yearp) {
7268: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7269: }
7270: } /* end i */
7271: if (h*hstepm/YEARM*stepm==yearp) {
7272: fprintf(ficresf," %.3f", ppij);
7273: }
7274: }/* end j */
7275: } /* end h */
7276: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7277: } /* end agec */
7278: } /* end yearp */
7279: } /* end k */
1.219 brouard 7280:
1.126 brouard 7281: fclose(ficresf);
1.215 brouard 7282: printf("End of Computing forecasting \n");
7283: fprintf(ficlog,"End of Computing forecasting\n");
7284:
1.126 brouard 7285: }
7286:
1.218 brouard 7287: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7288: /* 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 7289: /* /\* back1, year, month, day of starting backection */
7290: /* agemin, agemax range of age */
7291: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7292: /* anback2 year of en of backection (same day and month as back1). */
7293: /* *\/ */
7294: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7295: /* double agec; /\* generic age *\/ */
7296: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7297: /* double *popeffectif,*popcount; */
7298: /* double ***p3mat; */
7299: /* /\* double ***mobaverage; *\/ */
7300: /* char fileresfb[FILENAMELENGTH]; */
7301:
7302: /* agelim=AGESUP; */
7303: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7304: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7305: /* We still use firstpass and lastpass as another selection. */
7306: /* *\/ */
7307: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7308: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7309: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7310:
7311: /* strcpy(fileresfb,"FB_"); */
7312: /* strcat(fileresfb,fileresu); */
7313: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7314: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7315: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7316: /* } */
7317: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7318: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7319:
1.225 brouard 7320: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7321:
7322: /* /\* if (mobilav!=0) { *\/ */
7323: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7324: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7325: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7326: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7327: /* /\* } *\/ */
7328: /* /\* } *\/ */
7329:
7330: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7331: /* if (stepm<=12) stepsize=1; */
7332: /* if(estepm < stepm){ */
7333: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7334: /* } */
7335: /* else hstepm=estepm; */
7336:
7337: /* hstepm=hstepm/stepm; */
7338: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7339: /* fractional in yp1 *\/ */
7340: /* anprojmean=yp; */
7341: /* yp2=modf((yp1*12),&yp); */
7342: /* mprojmean=yp; */
7343: /* yp1=modf((yp2*30.5),&yp); */
7344: /* jprojmean=yp; */
7345: /* if(jprojmean==0) jprojmean=1; */
7346: /* if(mprojmean==0) jprojmean=1; */
7347:
1.225 brouard 7348: /* i1=cptcoveff; */
1.218 brouard 7349: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7350:
1.218 brouard 7351: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7352:
1.218 brouard 7353: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7354:
7355: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7356: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7357: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7358: /* k=k+1; */
7359: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7360: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7361: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7362: /* } */
7363: /* fprintf(ficresfb," yearbproj age"); */
7364: /* for(j=1; j<=nlstate+ndeath;j++){ */
7365: /* for(i=1; i<=nlstate;i++) */
7366: /* fprintf(ficresfb," p%d%d",i,j); */
7367: /* fprintf(ficresfb," p.%d",j); */
7368: /* } */
7369: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7370: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7371: /* fprintf(ficresfb,"\n"); */
7372: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7373: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7374: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7375: /* nhstepm = nhstepm/hstepm; */
7376: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7377: /* oldm=oldms;savm=savms; */
7378: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7379: /* for (h=0; h<=nhstepm; h++){ */
7380: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7381: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7382: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7383: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7384: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7385: /* } */
7386: /* for(j=1; j<=nlstate+ndeath;j++) { */
7387: /* ppij=0.; */
7388: /* for(i=1; i<=nlstate;i++) { */
7389: /* if (mobilav==1) */
7390: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7391: /* else { */
7392: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7393: /* } */
7394: /* if (h*hstepm/YEARM*stepm== yearp) { */
7395: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7396: /* } */
7397: /* } /\* end i *\/ */
7398: /* if (h*hstepm/YEARM*stepm==yearp) { */
7399: /* fprintf(ficresfb," %.3f", ppij); */
7400: /* } */
7401: /* }/\* end j *\/ */
7402: /* } /\* end h *\/ */
7403: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7404: /* } /\* end agec *\/ */
7405: /* } /\* end yearp *\/ */
7406: /* } /\* end cptcod *\/ */
7407: /* } /\* end cptcov *\/ */
7408:
7409: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7410:
7411: /* fclose(ficresfb); */
7412: /* printf("End of Computing Back forecasting \n"); */
7413: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7414:
1.218 brouard 7415: /* } */
1.217 brouard 7416:
1.126 brouard 7417: /************** Forecasting *****not tested NB*************/
1.227 brouard 7418: /* 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 7419:
1.227 brouard 7420: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7421: /* int *popage; */
7422: /* double calagedatem, agelim, kk1, kk2; */
7423: /* double *popeffectif,*popcount; */
7424: /* double ***p3mat,***tabpop,***tabpopprev; */
7425: /* /\* double ***mobaverage; *\/ */
7426: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7427:
1.227 brouard 7428: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7429: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7430: /* agelim=AGESUP; */
7431: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7432:
1.227 brouard 7433: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7434:
7435:
1.227 brouard 7436: /* strcpy(filerespop,"POP_"); */
7437: /* strcat(filerespop,fileresu); */
7438: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7439: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7440: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7441: /* } */
7442: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7443: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7444:
1.227 brouard 7445: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7446:
1.227 brouard 7447: /* /\* if (mobilav!=0) { *\/ */
7448: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7449: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7450: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7451: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7452: /* /\* } *\/ */
7453: /* /\* } *\/ */
1.126 brouard 7454:
1.227 brouard 7455: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7456: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7457:
1.227 brouard 7458: /* agelim=AGESUP; */
1.126 brouard 7459:
1.227 brouard 7460: /* hstepm=1; */
7461: /* hstepm=hstepm/stepm; */
1.218 brouard 7462:
1.227 brouard 7463: /* if (popforecast==1) { */
7464: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7465: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7466: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7467: /* } */
7468: /* popage=ivector(0,AGESUP); */
7469: /* popeffectif=vector(0,AGESUP); */
7470: /* popcount=vector(0,AGESUP); */
1.126 brouard 7471:
1.227 brouard 7472: /* i=1; */
7473: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7474:
1.227 brouard 7475: /* imx=i; */
7476: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7477: /* } */
1.218 brouard 7478:
1.227 brouard 7479: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7480: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7481: /* k=k+1; */
7482: /* fprintf(ficrespop,"\n#******"); */
7483: /* for(j=1;j<=cptcoveff;j++) { */
7484: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7485: /* } */
7486: /* fprintf(ficrespop,"******\n"); */
7487: /* fprintf(ficrespop,"# Age"); */
7488: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7489: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7490:
1.227 brouard 7491: /* for (cpt=0; cpt<=0;cpt++) { */
7492: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7493:
1.227 brouard 7494: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7495: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7496: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7497:
1.227 brouard 7498: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7499: /* oldm=oldms;savm=savms; */
7500: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7501:
1.227 brouard 7502: /* for (h=0; h<=nhstepm; h++){ */
7503: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7504: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7505: /* } */
7506: /* for(j=1; j<=nlstate+ndeath;j++) { */
7507: /* kk1=0.;kk2=0; */
7508: /* for(i=1; i<=nlstate;i++) { */
7509: /* if (mobilav==1) */
7510: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7511: /* else { */
7512: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7513: /* } */
7514: /* } */
7515: /* if (h==(int)(calagedatem+12*cpt)){ */
7516: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7517: /* /\*fprintf(ficrespop," %.3f", kk1); */
7518: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7519: /* } */
7520: /* } */
7521: /* for(i=1; i<=nlstate;i++){ */
7522: /* kk1=0.; */
7523: /* for(j=1; j<=nlstate;j++){ */
7524: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7525: /* } */
7526: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7527: /* } */
1.218 brouard 7528:
1.227 brouard 7529: /* if (h==(int)(calagedatem+12*cpt)) */
7530: /* for(j=1; j<=nlstate;j++) */
7531: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7532: /* } */
7533: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7534: /* } */
7535: /* } */
1.218 brouard 7536:
1.227 brouard 7537: /* /\******\/ */
1.218 brouard 7538:
1.227 brouard 7539: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7540: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7541: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7542: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7543: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7544:
1.227 brouard 7545: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7546: /* oldm=oldms;savm=savms; */
7547: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7548: /* for (h=0; h<=nhstepm; h++){ */
7549: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7550: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7551: /* } */
7552: /* for(j=1; j<=nlstate+ndeath;j++) { */
7553: /* kk1=0.;kk2=0; */
7554: /* for(i=1; i<=nlstate;i++) { */
7555: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7556: /* } */
7557: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7558: /* } */
7559: /* } */
7560: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7561: /* } */
7562: /* } */
7563: /* } */
7564: /* } */
1.218 brouard 7565:
1.227 brouard 7566: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7567:
1.227 brouard 7568: /* if (popforecast==1) { */
7569: /* free_ivector(popage,0,AGESUP); */
7570: /* free_vector(popeffectif,0,AGESUP); */
7571: /* free_vector(popcount,0,AGESUP); */
7572: /* } */
7573: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7574: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7575: /* fclose(ficrespop); */
7576: /* } /\* End of popforecast *\/ */
1.218 brouard 7577:
1.126 brouard 7578: int fileappend(FILE *fichier, char *optionfich)
7579: {
7580: if((fichier=fopen(optionfich,"a"))==NULL) {
7581: printf("Problem with file: %s\n", optionfich);
7582: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7583: return (0);
7584: }
7585: fflush(fichier);
7586: return (1);
7587: }
7588:
7589:
7590: /**************** function prwizard **********************/
7591: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7592: {
7593:
7594: /* Wizard to print covariance matrix template */
7595:
1.164 brouard 7596: char ca[32], cb[32];
7597: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7598: int numlinepar;
7599:
7600: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7601: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7602: for(i=1; i <=nlstate; i++){
7603: jj=0;
7604: for(j=1; j <=nlstate+ndeath; j++){
7605: if(j==i) continue;
7606: jj++;
7607: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7608: printf("%1d%1d",i,j);
7609: fprintf(ficparo,"%1d%1d",i,j);
7610: for(k=1; k<=ncovmodel;k++){
7611: /* printf(" %lf",param[i][j][k]); */
7612: /* fprintf(ficparo," %lf",param[i][j][k]); */
7613: printf(" 0.");
7614: fprintf(ficparo," 0.");
7615: }
7616: printf("\n");
7617: fprintf(ficparo,"\n");
7618: }
7619: }
7620: printf("# Scales (for hessian or gradient estimation)\n");
7621: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7622: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7623: for(i=1; i <=nlstate; i++){
7624: jj=0;
7625: for(j=1; j <=nlstate+ndeath; j++){
7626: if(j==i) continue;
7627: jj++;
7628: fprintf(ficparo,"%1d%1d",i,j);
7629: printf("%1d%1d",i,j);
7630: fflush(stdout);
7631: for(k=1; k<=ncovmodel;k++){
7632: /* printf(" %le",delti3[i][j][k]); */
7633: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7634: printf(" 0.");
7635: fprintf(ficparo," 0.");
7636: }
7637: numlinepar++;
7638: printf("\n");
7639: fprintf(ficparo,"\n");
7640: }
7641: }
7642: printf("# Covariance matrix\n");
7643: /* # 121 Var(a12)\n\ */
7644: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7645: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7646: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7647: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7648: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7649: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7650: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7651: fflush(stdout);
7652: fprintf(ficparo,"# Covariance matrix\n");
7653: /* # 121 Var(a12)\n\ */
7654: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7655: /* # ...\n\ */
7656: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7657:
7658: for(itimes=1;itimes<=2;itimes++){
7659: jj=0;
7660: for(i=1; i <=nlstate; i++){
7661: for(j=1; j <=nlstate+ndeath; j++){
7662: if(j==i) continue;
7663: for(k=1; k<=ncovmodel;k++){
7664: jj++;
7665: ca[0]= k+'a'-1;ca[1]='\0';
7666: if(itimes==1){
7667: printf("#%1d%1d%d",i,j,k);
7668: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7669: }else{
7670: printf("%1d%1d%d",i,j,k);
7671: fprintf(ficparo,"%1d%1d%d",i,j,k);
7672: /* printf(" %.5le",matcov[i][j]); */
7673: }
7674: ll=0;
7675: for(li=1;li <=nlstate; li++){
7676: for(lj=1;lj <=nlstate+ndeath; lj++){
7677: if(lj==li) continue;
7678: for(lk=1;lk<=ncovmodel;lk++){
7679: ll++;
7680: if(ll<=jj){
7681: cb[0]= lk +'a'-1;cb[1]='\0';
7682: if(ll<jj){
7683: if(itimes==1){
7684: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7685: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7686: }else{
7687: printf(" 0.");
7688: fprintf(ficparo," 0.");
7689: }
7690: }else{
7691: if(itimes==1){
7692: printf(" Var(%s%1d%1d)",ca,i,j);
7693: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
7694: }else{
7695: printf(" 0.");
7696: fprintf(ficparo," 0.");
7697: }
7698: }
7699: }
7700: } /* end lk */
7701: } /* end lj */
7702: } /* end li */
7703: printf("\n");
7704: fprintf(ficparo,"\n");
7705: numlinepar++;
7706: } /* end k*/
7707: } /*end j */
7708: } /* end i */
7709: } /* end itimes */
7710:
7711: } /* end of prwizard */
7712: /******************* Gompertz Likelihood ******************************/
7713: double gompertz(double x[])
7714: {
7715: double A,B,L=0.0,sump=0.,num=0.;
7716: int i,n=0; /* n is the size of the sample */
7717:
1.220 brouard 7718: for (i=1;i<=imx ; i++) {
1.126 brouard 7719: sump=sump+weight[i];
7720: /* sump=sump+1;*/
7721: num=num+1;
7722: }
7723:
7724:
7725: /* for (i=0; i<=imx; i++)
7726: 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]);*/
7727:
7728: for (i=1;i<=imx ; i++)
7729: {
7730: if (cens[i] == 1 && wav[i]>1)
7731: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
7732:
7733: if (cens[i] == 0 && wav[i]>1)
7734: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
7735: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
7736:
7737: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7738: if (wav[i] > 1 ) { /* ??? */
7739: L=L+A*weight[i];
7740: /* 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]);*/
7741: }
7742: }
7743:
7744: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7745:
7746: return -2*L*num/sump;
7747: }
7748:
1.136 brouard 7749: #ifdef GSL
7750: /******************* Gompertz_f Likelihood ******************************/
7751: double gompertz_f(const gsl_vector *v, void *params)
7752: {
7753: double A,B,LL=0.0,sump=0.,num=0.;
7754: double *x= (double *) v->data;
7755: int i,n=0; /* n is the size of the sample */
7756:
7757: for (i=0;i<=imx-1 ; i++) {
7758: sump=sump+weight[i];
7759: /* sump=sump+1;*/
7760: num=num+1;
7761: }
7762:
7763:
7764: /* for (i=0; i<=imx; i++)
7765: 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]);*/
7766: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
7767: for (i=1;i<=imx ; i++)
7768: {
7769: if (cens[i] == 1 && wav[i]>1)
7770: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
7771:
7772: if (cens[i] == 0 && wav[i]>1)
7773: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
7774: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
7775:
7776: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7777: if (wav[i] > 1 ) { /* ??? */
7778: LL=LL+A*weight[i];
7779: /* 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]);*/
7780: }
7781: }
7782:
7783: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7784: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
7785:
7786: return -2*LL*num/sump;
7787: }
7788: #endif
7789:
1.126 brouard 7790: /******************* Printing html file ***********/
1.201 brouard 7791: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7792: int lastpass, int stepm, int weightopt, char model[],\
7793: int imx, double p[],double **matcov,double agemortsup){
7794: int i,k;
7795:
7796: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
7797: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
7798: for (i=1;i<=2;i++)
7799: 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 7800: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 7801: fprintf(fichtm,"</ul>");
7802:
7803: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
7804:
7805: 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>");
7806:
7807: for (k=agegomp;k<(agemortsup-2);k++)
7808: 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]);
7809:
7810:
7811: fflush(fichtm);
7812: }
7813:
7814: /******************* Gnuplot file **************/
1.201 brouard 7815: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 7816:
7817: char dirfileres[132],optfileres[132];
1.164 brouard 7818:
1.126 brouard 7819: int ng;
7820:
7821:
7822: /*#ifdef windows */
7823: fprintf(ficgp,"cd \"%s\" \n",pathc);
7824: /*#endif */
7825:
7826:
7827: strcpy(dirfileres,optionfilefiname);
7828: strcpy(optfileres,"vpl");
1.199 brouard 7829: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 7830: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 7831: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 7832: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 7833: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
7834:
7835: }
7836:
1.136 brouard 7837: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
7838: {
1.126 brouard 7839:
1.136 brouard 7840: /*-------- data file ----------*/
7841: FILE *fic;
7842: char dummy[]=" ";
1.240 brouard 7843: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 7844: int lstra;
1.136 brouard 7845: int linei, month, year,iout;
7846: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 7847: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 7848: char *stratrunc;
1.223 brouard 7849:
1.240 brouard 7850: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
7851: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 7852:
1.240 brouard 7853: for(v=1; v <=ncovcol;v++){
7854: DummyV[v]=0;
7855: FixedV[v]=0;
7856: }
7857: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
7858: DummyV[v]=1;
7859: FixedV[v]=0;
7860: }
7861: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
7862: DummyV[v]=0;
7863: FixedV[v]=1;
7864: }
7865: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
7866: DummyV[v]=1;
7867: FixedV[v]=1;
7868: }
7869: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
7870: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
7871: 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]);
7872: }
1.126 brouard 7873:
1.136 brouard 7874: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 7875: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
7876: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 7877: }
1.126 brouard 7878:
1.136 brouard 7879: i=1;
7880: linei=0;
7881: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
7882: linei=linei+1;
7883: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
7884: if(line[j] == '\t')
7885: line[j] = ' ';
7886: }
7887: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
7888: ;
7889: };
7890: line[j+1]=0; /* Trims blanks at end of line */
7891: if(line[0]=='#'){
7892: fprintf(ficlog,"Comment line\n%s\n",line);
7893: printf("Comment line\n%s\n",line);
7894: continue;
7895: }
7896: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 7897: strcpy(line, linetmp);
1.223 brouard 7898:
7899: /* Loops on waves */
7900: for (j=maxwav;j>=1;j--){
7901: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 7902: cutv(stra, strb, line, ' ');
7903: if(strb[0]=='.') { /* Missing value */
7904: lval=-1;
7905: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
7906: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
7907: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
7908: 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);
7909: 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);
7910: return 1;
7911: }
7912: }else{
7913: errno=0;
7914: /* what_kind_of_number(strb); */
7915: dval=strtod(strb,&endptr);
7916: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
7917: /* if(strb != endptr && *endptr == '\0') */
7918: /* dval=dlval; */
7919: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
7920: if( strb[0]=='\0' || (*endptr != '\0')){
7921: 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);
7922: 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);
7923: return 1;
7924: }
7925: cotqvar[j][iv][i]=dval;
7926: cotvar[j][ntv+iv][i]=dval;
7927: }
7928: strcpy(line,stra);
1.223 brouard 7929: }/* end loop ntqv */
1.225 brouard 7930:
1.223 brouard 7931: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 7932: cutv(stra, strb, line, ' ');
7933: if(strb[0]=='.') { /* Missing value */
7934: lval=-1;
7935: }else{
7936: errno=0;
7937: lval=strtol(strb,&endptr,10);
7938: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
7939: if( strb[0]=='\0' || (*endptr != '\0')){
7940: 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);
7941: 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);
7942: return 1;
7943: }
7944: }
7945: if(lval <-1 || lval >1){
7946: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 7947: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
7948: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 7949: For example, for multinomial values like 1, 2 and 3,\n \
7950: build V1=0 V2=0 for the reference value (1),\n \
7951: V1=1 V2=0 for (2) \n \
1.223 brouard 7952: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 7953: output of IMaCh is often meaningless.\n \
1.223 brouard 7954: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 7955: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 7956: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
7957: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 7958: For example, for multinomial values like 1, 2 and 3,\n \
7959: build V1=0 V2=0 for the reference value (1),\n \
7960: V1=1 V2=0 for (2) \n \
1.223 brouard 7961: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 7962: output of IMaCh is often meaningless.\n \
1.223 brouard 7963: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 7964: return 1;
7965: }
7966: cotvar[j][iv][i]=(double)(lval);
7967: strcpy(line,stra);
1.223 brouard 7968: }/* end loop ntv */
1.225 brouard 7969:
1.223 brouard 7970: /* Statuses at wave */
1.137 brouard 7971: cutv(stra, strb, line, ' ');
1.223 brouard 7972: if(strb[0]=='.') { /* Missing value */
1.238 brouard 7973: lval=-1;
1.136 brouard 7974: }else{
1.238 brouard 7975: errno=0;
7976: lval=strtol(strb,&endptr,10);
7977: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
7978: if( strb[0]=='\0' || (*endptr != '\0')){
7979: 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);
7980: 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);
7981: return 1;
7982: }
1.136 brouard 7983: }
1.225 brouard 7984:
1.136 brouard 7985: s[j][i]=lval;
1.225 brouard 7986:
1.223 brouard 7987: /* Date of Interview */
1.136 brouard 7988: strcpy(line,stra);
7989: cutv(stra, strb,line,' ');
1.169 brouard 7990: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 7991: }
1.169 brouard 7992: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 7993: month=99;
7994: year=9999;
1.136 brouard 7995: }else{
1.225 brouard 7996: 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);
7997: 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);
7998: return 1;
1.136 brouard 7999: }
8000: anint[j][i]= (double) year;
8001: mint[j][i]= (double)month;
8002: strcpy(line,stra);
1.223 brouard 8003: } /* End loop on waves */
1.225 brouard 8004:
1.223 brouard 8005: /* Date of death */
1.136 brouard 8006: cutv(stra, strb,line,' ');
1.169 brouard 8007: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8008: }
1.169 brouard 8009: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8010: month=99;
8011: year=9999;
8012: }else{
1.141 brouard 8013: 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 8014: 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);
8015: return 1;
1.136 brouard 8016: }
8017: andc[i]=(double) year;
8018: moisdc[i]=(double) month;
8019: strcpy(line,stra);
8020:
1.223 brouard 8021: /* Date of birth */
1.136 brouard 8022: cutv(stra, strb,line,' ');
1.169 brouard 8023: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8024: }
1.169 brouard 8025: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8026: month=99;
8027: year=9999;
8028: }else{
1.141 brouard 8029: 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);
8030: 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 8031: return 1;
1.136 brouard 8032: }
8033: if (year==9999) {
1.141 brouard 8034: 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);
8035: 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 8036: return 1;
8037:
1.136 brouard 8038: }
8039: annais[i]=(double)(year);
8040: moisnais[i]=(double)(month);
8041: strcpy(line,stra);
1.225 brouard 8042:
1.223 brouard 8043: /* Sample weight */
1.136 brouard 8044: cutv(stra, strb,line,' ');
8045: errno=0;
8046: dval=strtod(strb,&endptr);
8047: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8048: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8049: 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 8050: fflush(ficlog);
8051: return 1;
8052: }
8053: weight[i]=dval;
8054: strcpy(line,stra);
1.225 brouard 8055:
1.223 brouard 8056: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8057: cutv(stra, strb, line, ' ');
8058: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8059: lval=-1;
1.223 brouard 8060: }else{
1.225 brouard 8061: errno=0;
8062: /* what_kind_of_number(strb); */
8063: dval=strtod(strb,&endptr);
8064: /* if(strb != endptr && *endptr == '\0') */
8065: /* dval=dlval; */
8066: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8067: if( strb[0]=='\0' || (*endptr != '\0')){
8068: 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);
8069: 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);
8070: return 1;
8071: }
8072: coqvar[iv][i]=dval;
1.226 brouard 8073: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8074: }
8075: strcpy(line,stra);
8076: }/* end loop nqv */
1.136 brouard 8077:
1.223 brouard 8078: /* Covariate values */
1.136 brouard 8079: for (j=ncovcol;j>=1;j--){
8080: cutv(stra, strb,line,' ');
1.223 brouard 8081: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8082: lval=-1;
1.136 brouard 8083: }else{
1.225 brouard 8084: errno=0;
8085: lval=strtol(strb,&endptr,10);
8086: if( strb[0]=='\0' || (*endptr != '\0')){
8087: 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);
8088: 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);
8089: return 1;
8090: }
1.136 brouard 8091: }
8092: if(lval <-1 || lval >1){
1.225 brouard 8093: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8094: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8095: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8096: For example, for multinomial values like 1, 2 and 3,\n \
8097: build V1=0 V2=0 for the reference value (1),\n \
8098: V1=1 V2=0 for (2) \n \
1.136 brouard 8099: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8100: output of IMaCh is often meaningless.\n \
1.136 brouard 8101: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8102: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8103: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8104: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8105: For example, for multinomial values like 1, 2 and 3,\n \
8106: build V1=0 V2=0 for the reference value (1),\n \
8107: V1=1 V2=0 for (2) \n \
1.136 brouard 8108: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8109: output of IMaCh is often meaningless.\n \
1.136 brouard 8110: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8111: return 1;
1.136 brouard 8112: }
8113: covar[j][i]=(double)(lval);
8114: strcpy(line,stra);
8115: }
8116: lstra=strlen(stra);
1.225 brouard 8117:
1.136 brouard 8118: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8119: stratrunc = &(stra[lstra-9]);
8120: num[i]=atol(stratrunc);
8121: }
8122: else
8123: num[i]=atol(stra);
8124: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8125: 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;}*/
8126:
8127: i=i+1;
8128: } /* End loop reading data */
1.225 brouard 8129:
1.136 brouard 8130: *imax=i-1; /* Number of individuals */
8131: fclose(fic);
1.225 brouard 8132:
1.136 brouard 8133: return (0);
1.164 brouard 8134: /* endread: */
1.225 brouard 8135: printf("Exiting readdata: ");
8136: fclose(fic);
8137: return (1);
1.223 brouard 8138: }
1.126 brouard 8139:
1.234 brouard 8140: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8141: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8142: while (*p2 == ' ')
1.234 brouard 8143: p2++;
8144: /* while ((*p1++ = *p2++) !=0) */
8145: /* ; */
8146: /* do */
8147: /* while (*p2 == ' ') */
8148: /* p2++; */
8149: /* while (*p1++ == *p2++); */
8150: *stri=p2;
1.145 brouard 8151: }
8152:
1.235 brouard 8153: int decoderesult ( char resultline[], int nres)
1.230 brouard 8154: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8155: {
1.235 brouard 8156: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8157: char resultsav[MAXLINE];
1.234 brouard 8158: int resultmodel[MAXLINE];
8159: int modelresult[MAXLINE];
1.230 brouard 8160: char stra[80], strb[80], strc[80], strd[80],stre[80];
8161:
1.234 brouard 8162: removefirstspace(&resultline);
1.233 brouard 8163: printf("decoderesult:%s\n",resultline);
1.230 brouard 8164:
8165: if (strstr(resultline,"v") !=0){
8166: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8167: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8168: return 1;
8169: }
8170: trimbb(resultsav, resultline);
8171: if (strlen(resultsav) >1){
8172: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8173: }
1.234 brouard 8174: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8175: 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);
8176: 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);
8177: }
8178: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8179: if(nbocc(resultsav,'=') >1){
8180: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8181: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8182: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8183: }else
8184: cutl(strc,strd,resultsav,'=');
1.230 brouard 8185: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8186:
1.230 brouard 8187: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8188: Tvarsel[k]=atoi(strc);
8189: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8190: /* cptcovsel++; */
8191: if (nbocc(stra,'=') >0)
8192: strcpy(resultsav,stra); /* and analyzes it */
8193: }
1.235 brouard 8194: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8195: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8196: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8197: match=0;
1.236 brouard 8198: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8199: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8200: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8201: match=1;
8202: break;
8203: }
8204: }
8205: if(match == 0){
8206: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8207: }
8208: }
8209: }
1.235 brouard 8210: /* Checking for missing or useless values in comparison of current model needs */
8211: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8212: match=0;
1.235 brouard 8213: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8214: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8215: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8216: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8217: ++match;
8218: }
8219: }
8220: }
8221: if(match == 0){
8222: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8223: }else if(match > 1){
8224: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8225: }
8226: }
1.235 brouard 8227:
1.234 brouard 8228: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8229: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8230: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8231: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8232: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8233: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8234: /* 1 0 0 0 */
8235: /* 2 1 0 0 */
8236: /* 3 0 1 0 */
8237: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8238: /* 5 0 0 1 */
8239: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8240: /* 7 0 1 1 */
8241: /* 8 1 1 1 */
1.237 brouard 8242: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8243: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8244: /* V5*age V5 known which value for nres? */
8245: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8246: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8247: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8248: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8249: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8250: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8251: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8252: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8253: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8254: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8255: k4++;;
8256: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8257: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8258: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8259: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8260: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8261: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8262: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8263: k4q++;;
8264: }
8265: }
1.234 brouard 8266:
1.235 brouard 8267: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8268: return (0);
8269: }
1.235 brouard 8270:
1.230 brouard 8271: int decodemodel( char model[], int lastobs)
8272: /**< This routine decodes the model and returns:
1.224 brouard 8273: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8274: * - nagesqr = 1 if age*age in the model, otherwise 0.
8275: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8276: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8277: * - cptcovage number of covariates with age*products =2
8278: * - cptcovs number of simple covariates
8279: * - 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
8280: * which is a new column after the 9 (ncovcol) variables.
8281: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8282: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8283: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8284: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8285: */
1.136 brouard 8286: {
1.238 brouard 8287: int i, j, k, ks, v;
1.227 brouard 8288: int j1, k1, k2, k3, k4;
1.136 brouard 8289: char modelsav[80];
1.145 brouard 8290: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8291: char *strpt;
1.136 brouard 8292:
1.145 brouard 8293: /*removespace(model);*/
1.136 brouard 8294: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8295: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8296: if (strstr(model,"AGE") !=0){
1.192 brouard 8297: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8298: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8299: return 1;
8300: }
1.141 brouard 8301: if (strstr(model,"v") !=0){
8302: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8303: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8304: return 1;
8305: }
1.187 brouard 8306: strcpy(modelsav,model);
8307: if ((strpt=strstr(model,"age*age")) !=0){
8308: printf(" strpt=%s, model=%s\n",strpt, model);
8309: if(strpt != model){
1.234 brouard 8310: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8311: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8312: corresponding column of parameters.\n",model);
1.234 brouard 8313: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8314: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8315: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8316: return 1;
1.225 brouard 8317: }
1.187 brouard 8318: nagesqr=1;
8319: if (strstr(model,"+age*age") !=0)
1.234 brouard 8320: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8321: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8322: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8323: else
1.234 brouard 8324: substrchaine(modelsav, model, "age*age");
1.187 brouard 8325: }else
8326: nagesqr=0;
8327: if (strlen(modelsav) >1){
8328: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8329: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8330: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8331: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8332: * cst, age and age*age
8333: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8334: /* including age products which are counted in cptcovage.
8335: * but the covariates which are products must be treated
8336: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8337: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8338: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8339:
8340:
1.187 brouard 8341: /* Design
8342: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8343: * < ncovcol=8 >
8344: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8345: * k= 1 2 3 4 5 6 7 8
8346: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8347: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8348: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8349: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8350: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8351: * Tage[++cptcovage]=k
8352: * if products, new covar are created after ncovcol with k1
8353: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8354: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8355: * 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
8356: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8357: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8358: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8359: * < ncovcol=8 >
8360: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8361: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8362: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8363: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8364: * p Tprod[1]@2={ 6, 5}
8365: *p Tvard[1][1]@4= {7, 8, 5, 6}
8366: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8367: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8368: *How to reorganize?
8369: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8370: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8371: * {2, 1, 4, 8, 5, 6, 3, 7}
8372: * Struct []
8373: */
1.225 brouard 8374:
1.187 brouard 8375: /* This loop fills the array Tvar from the string 'model'.*/
8376: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8377: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8378: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8379: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8380: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8381: /* k=1 Tvar[1]=2 (from V2) */
8382: /* k=5 Tvar[5] */
8383: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8384: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8385: /* } */
1.198 brouard 8386: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8387: /*
8388: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8389: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8390: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8391: }
1.187 brouard 8392: cptcovage=0;
8393: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8394: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8395: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8396: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8397: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8398: /*scanf("%d",i);*/
8399: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8400: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8401: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8402: /* covar is not filled and then is empty */
8403: cptcovprod--;
8404: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8405: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8406: Typevar[k]=1; /* 1 for age product */
8407: cptcovage++; /* Sums the number of covariates which include age as a product */
8408: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8409: /*printf("stre=%s ", stre);*/
8410: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8411: cptcovprod--;
8412: cutl(stre,strb,strc,'V');
8413: Tvar[k]=atoi(stre);
8414: Typevar[k]=1; /* 1 for age product */
8415: cptcovage++;
8416: Tage[cptcovage]=k;
8417: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8418: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8419: cptcovn++;
8420: cptcovprodnoage++;k1++;
8421: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8422: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8423: because this model-covariate is a construction we invent a new column
8424: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8425: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8426: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8427: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8428: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8429: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8430: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8431: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8432: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8433: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8434: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8435: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8436: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8437: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8438: for (i=1; i<=lastobs;i++){
8439: /* Computes the new covariate which is a product of
8440: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8441: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8442: }
8443: } /* End age is not in the model */
8444: } /* End if model includes a product */
8445: else { /* no more sum */
8446: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8447: /* scanf("%d",i);*/
8448: cutl(strd,strc,strb,'V');
8449: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8450: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8451: Tvar[k]=atoi(strd);
8452: Typevar[k]=0; /* 0 for simple covariates */
8453: }
8454: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8455: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8456: scanf("%d",i);*/
1.187 brouard 8457: } /* end of loop + on total covariates */
8458: } /* end if strlen(modelsave == 0) age*age might exist */
8459: } /* end if strlen(model == 0) */
1.136 brouard 8460:
8461: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8462: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8463:
1.136 brouard 8464: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8465: printf("cptcovprod=%d ", cptcovprod);
8466: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8467: scanf("%d ",i);*/
8468:
8469:
1.230 brouard 8470: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8471: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8472: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8473: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8474: k = 1 2 3 4 5 6 7 8 9
8475: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8476: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8477: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8478: Dummy[k] 1 0 0 0 3 1 1 2 3
8479: Tmodelind[combination of covar]=k;
1.225 brouard 8480: */
8481: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8482: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8483: /* 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 8484: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8485: printf("Model=%s\n\
8486: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8487: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8488: 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);
8489: fprintf(ficlog,"Model=%s\n\
8490: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8491: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8492: 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 8493: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8494: 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 */
8495: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8496: Fixed[k]= 0;
8497: Dummy[k]= 0;
1.225 brouard 8498: ncoveff++;
1.232 brouard 8499: ncovf++;
1.234 brouard 8500: nsd++;
8501: modell[k].maintype= FTYPE;
8502: TvarsD[nsd]=Tvar[k];
8503: TvarsDind[nsd]=k;
8504: TvarF[ncovf]=Tvar[k];
8505: TvarFind[ncovf]=k;
8506: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8507: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8508: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8509: Fixed[k]= 0;
8510: Dummy[k]= 0;
8511: ncoveff++;
8512: ncovf++;
8513: modell[k].maintype= FTYPE;
8514: TvarF[ncovf]=Tvar[k];
8515: TvarFind[ncovf]=k;
1.230 brouard 8516: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8517: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8518: }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 8519: Fixed[k]= 0;
8520: Dummy[k]= 1;
1.230 brouard 8521: nqfveff++;
1.234 brouard 8522: modell[k].maintype= FTYPE;
8523: modell[k].subtype= FQ;
8524: nsq++;
8525: TvarsQ[nsq]=Tvar[k];
8526: TvarsQind[nsq]=k;
1.232 brouard 8527: ncovf++;
1.234 brouard 8528: TvarF[ncovf]=Tvar[k];
8529: TvarFind[ncovf]=k;
1.231 brouard 8530: 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 8531: 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 */
1.234 brouard 8532: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying variables */
1.227 brouard 8533: Fixed[k]= 1;
8534: Dummy[k]= 0;
1.225 brouard 8535: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8536: modell[k].maintype= VTYPE;
8537: modell[k].subtype= VD;
8538: nsd++;
8539: TvarsD[nsd]=Tvar[k];
8540: TvarsDind[nsd]=k;
8541: ncovv++; /* Only simple time varying variables */
8542: TvarV[ncovv]=Tvar[k];
8543: TvarVind[ncovv]=k;
1.231 brouard 8544: 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 */
8545: 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 8546: 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);
8547: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8548: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8549: Fixed[k]= 1;
8550: Dummy[k]= 1;
8551: nqtveff++;
8552: modell[k].maintype= VTYPE;
8553: modell[k].subtype= VQ;
8554: ncovv++; /* Only simple time varying variables */
8555: nsq++;
8556: TvarsQ[nsq]=Tvar[k];
8557: TvarsQind[nsq]=k;
8558: TvarV[ncovv]=Tvar[k];
8559: TvarVind[ncovv]=k;
1.231 brouard 8560: 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 */
8561: 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 8562: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8563: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8564: 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 8565: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8566: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8567: ncova++;
8568: TvarA[ncova]=Tvar[k];
8569: TvarAind[ncova]=k;
1.231 brouard 8570: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8571: Fixed[k]= 2;
8572: Dummy[k]= 2;
8573: modell[k].maintype= ATYPE;
8574: modell[k].subtype= APFD;
8575: /* ncoveff++; */
1.227 brouard 8576: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8577: Fixed[k]= 2;
8578: Dummy[k]= 3;
8579: modell[k].maintype= ATYPE;
8580: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8581: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8582: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8583: Fixed[k]= 3;
8584: Dummy[k]= 2;
8585: modell[k].maintype= ATYPE;
8586: modell[k].subtype= APVD; /* Product age * varying dummy */
8587: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8588: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8589: Fixed[k]= 3;
8590: Dummy[k]= 3;
8591: modell[k].maintype= ATYPE;
8592: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8593: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8594: }
8595: }else if (Typevar[k] == 2) { /* product without age */
8596: k1=Tposprod[k];
8597: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8598: if(Tvard[k1][2] <=ncovcol){
8599: Fixed[k]= 1;
8600: Dummy[k]= 0;
8601: modell[k].maintype= FTYPE;
8602: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8603: ncovf++; /* Fixed variables without age */
8604: TvarF[ncovf]=Tvar[k];
8605: TvarFind[ncovf]=k;
8606: }else if(Tvard[k1][2] <=ncovcol+nqv){
8607: Fixed[k]= 0; /* or 2 ?*/
8608: Dummy[k]= 1;
8609: modell[k].maintype= FTYPE;
8610: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8611: ncovf++; /* Varying variables without age */
8612: TvarF[ncovf]=Tvar[k];
8613: TvarFind[ncovf]=k;
8614: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8615: Fixed[k]= 1;
8616: Dummy[k]= 0;
8617: modell[k].maintype= VTYPE;
8618: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8619: ncovv++; /* Varying variables without age */
8620: TvarV[ncovv]=Tvar[k];
8621: TvarVind[ncovv]=k;
8622: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8623: Fixed[k]= 1;
8624: Dummy[k]= 1;
8625: modell[k].maintype= VTYPE;
8626: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8627: ncovv++; /* Varying variables without age */
8628: TvarV[ncovv]=Tvar[k];
8629: TvarVind[ncovv]=k;
8630: }
1.227 brouard 8631: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8632: if(Tvard[k1][2] <=ncovcol){
8633: Fixed[k]= 0; /* or 2 ?*/
8634: Dummy[k]= 1;
8635: modell[k].maintype= FTYPE;
8636: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8637: ncovf++; /* Fixed variables without age */
8638: TvarF[ncovf]=Tvar[k];
8639: TvarFind[ncovf]=k;
8640: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8641: Fixed[k]= 1;
8642: Dummy[k]= 1;
8643: modell[k].maintype= VTYPE;
8644: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8645: ncovv++; /* Varying variables without age */
8646: TvarV[ncovv]=Tvar[k];
8647: TvarVind[ncovv]=k;
8648: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8649: Fixed[k]= 1;
8650: Dummy[k]= 1;
8651: modell[k].maintype= VTYPE;
8652: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8653: ncovv++; /* Varying variables without age */
8654: TvarV[ncovv]=Tvar[k];
8655: TvarVind[ncovv]=k;
8656: ncovv++; /* Varying variables without age */
8657: TvarV[ncovv]=Tvar[k];
8658: TvarVind[ncovv]=k;
8659: }
1.227 brouard 8660: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8661: if(Tvard[k1][2] <=ncovcol){
8662: Fixed[k]= 1;
8663: Dummy[k]= 1;
8664: modell[k].maintype= VTYPE;
8665: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8666: ncovv++; /* Varying variables without age */
8667: TvarV[ncovv]=Tvar[k];
8668: TvarVind[ncovv]=k;
8669: }else if(Tvard[k1][2] <=ncovcol+nqv){
8670: Fixed[k]= 1;
8671: Dummy[k]= 1;
8672: modell[k].maintype= VTYPE;
8673: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8674: ncovv++; /* Varying variables without age */
8675: TvarV[ncovv]=Tvar[k];
8676: TvarVind[ncovv]=k;
8677: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8678: Fixed[k]= 1;
8679: Dummy[k]= 0;
8680: modell[k].maintype= VTYPE;
8681: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8682: ncovv++; /* Varying variables without age */
8683: TvarV[ncovv]=Tvar[k];
8684: TvarVind[ncovv]=k;
8685: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8686: Fixed[k]= 1;
8687: Dummy[k]= 1;
8688: modell[k].maintype= VTYPE;
8689: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
8690: ncovv++; /* Varying variables without age */
8691: TvarV[ncovv]=Tvar[k];
8692: TvarVind[ncovv]=k;
8693: }
1.227 brouard 8694: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8695: if(Tvard[k1][2] <=ncovcol){
8696: Fixed[k]= 1;
8697: Dummy[k]= 1;
8698: modell[k].maintype= VTYPE;
8699: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
8700: ncovv++; /* Varying variables without age */
8701: TvarV[ncovv]=Tvar[k];
8702: TvarVind[ncovv]=k;
8703: }else if(Tvard[k1][2] <=ncovcol+nqv){
8704: Fixed[k]= 1;
8705: Dummy[k]= 1;
8706: modell[k].maintype= VTYPE;
8707: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
8708: ncovv++; /* Varying variables without age */
8709: TvarV[ncovv]=Tvar[k];
8710: TvarVind[ncovv]=k;
8711: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8712: Fixed[k]= 1;
8713: Dummy[k]= 1;
8714: modell[k].maintype= VTYPE;
8715: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
8716: ncovv++; /* Varying variables without age */
8717: TvarV[ncovv]=Tvar[k];
8718: TvarVind[ncovv]=k;
8719: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8720: Fixed[k]= 1;
8721: Dummy[k]= 1;
8722: modell[k].maintype= VTYPE;
8723: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
8724: ncovv++; /* Varying variables without age */
8725: TvarV[ncovv]=Tvar[k];
8726: TvarVind[ncovv]=k;
8727: }
1.227 brouard 8728: }else{
1.240 brouard 8729: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8730: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8731: } /*end k1*/
1.225 brouard 8732: }else{
1.226 brouard 8733: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
8734: 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 8735: }
1.227 brouard 8736: 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 8737: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 8738: 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]);
8739: }
8740: /* Searching for doublons in the model */
8741: for(k1=1; k1<= cptcovt;k1++){
8742: for(k2=1; k2 <k1;k2++){
8743: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 8744: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
8745: if(Tvar[k1]==Tvar[k2]){
8746: 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]]);
8747: 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);
8748: return(1);
8749: }
8750: }else if (Typevar[k1] ==2){
8751: k3=Tposprod[k1];
8752: k4=Tposprod[k2];
8753: 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])) ){
8754: 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]]);
8755: 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);
8756: return(1);
8757: }
8758: }
1.227 brouard 8759: }
8760: }
1.225 brouard 8761: }
8762: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
8763: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 8764: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
8765: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 8766: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 8767: /*endread:*/
1.225 brouard 8768: printf("Exiting decodemodel: ");
8769: return (1);
1.136 brouard 8770: }
8771:
1.169 brouard 8772: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.136 brouard 8773: {
8774: int i, m;
1.218 brouard 8775: int firstone=0;
8776:
1.136 brouard 8777: for (i=1; i<=imx; i++) {
8778: for(m=2; (m<= maxwav); m++) {
8779: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
8780: anint[m][i]=9999;
1.216 brouard 8781: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
8782: s[m][i]=-1;
1.136 brouard 8783: }
8784: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 8785: *nberr = *nberr + 1;
1.218 brouard 8786: if(firstone == 0){
8787: firstone=1;
8788: 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);
8789: }
8790: 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 8791: s[m][i]=-1;
8792: }
8793: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 8794: (*nberr)++;
1.136 brouard 8795: 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]);
8796: 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]);
8797: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
8798: }
8799: }
8800: }
8801:
8802: for (i=1; i<=imx; i++) {
8803: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
8804: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 8805: 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 8806: if (s[m][i] >= nlstate+1) {
1.169 brouard 8807: if(agedc[i]>0){
8808: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 8809: agev[m][i]=agedc[i];
1.214 brouard 8810: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 8811: }else {
1.136 brouard 8812: if ((int)andc[i]!=9999){
8813: nbwarn++;
8814: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
8815: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
8816: agev[m][i]=-1;
8817: }
8818: }
1.169 brouard 8819: } /* agedc > 0 */
1.214 brouard 8820: } /* end if */
1.136 brouard 8821: else if(s[m][i] !=9){ /* Standard case, age in fractional
8822: years but with the precision of a month */
8823: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
8824: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
8825: agev[m][i]=1;
8826: else if(agev[m][i] < *agemin){
8827: *agemin=agev[m][i];
8828: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
8829: }
8830: else if(agev[m][i] >*agemax){
8831: *agemax=agev[m][i];
1.156 brouard 8832: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 8833: }
8834: /*agev[m][i]=anint[m][i]-annais[i];*/
8835: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 8836: } /* en if 9*/
1.136 brouard 8837: else { /* =9 */
1.214 brouard 8838: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 8839: agev[m][i]=1;
8840: s[m][i]=-1;
8841: }
8842: }
1.214 brouard 8843: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 8844: agev[m][i]=1;
1.214 brouard 8845: else{
8846: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8847: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8848: agev[m][i]=0;
8849: }
8850: } /* End for lastpass */
8851: }
1.136 brouard 8852:
8853: for (i=1; i<=imx; i++) {
8854: for(m=firstpass; (m<=lastpass); m++){
8855: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 8856: (*nberr)++;
1.136 brouard 8857: 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);
8858: 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);
8859: return 1;
8860: }
8861: }
8862: }
8863:
8864: /*for (i=1; i<=imx; i++){
8865: for (m=firstpass; (m<lastpass); m++){
8866: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
8867: }
8868:
8869: }*/
8870:
8871:
1.139 brouard 8872: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
8873: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 8874:
8875: return (0);
1.164 brouard 8876: /* endread:*/
1.136 brouard 8877: printf("Exiting calandcheckages: ");
8878: return (1);
8879: }
8880:
1.172 brouard 8881: #if defined(_MSC_VER)
8882: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
8883: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
8884: //#include "stdafx.h"
8885: //#include <stdio.h>
8886: //#include <tchar.h>
8887: //#include <windows.h>
8888: //#include <iostream>
8889: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
8890:
8891: LPFN_ISWOW64PROCESS fnIsWow64Process;
8892:
8893: BOOL IsWow64()
8894: {
8895: BOOL bIsWow64 = FALSE;
8896:
8897: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
8898: // (HANDLE, PBOOL);
8899:
8900: //LPFN_ISWOW64PROCESS fnIsWow64Process;
8901:
8902: HMODULE module = GetModuleHandle(_T("kernel32"));
8903: const char funcName[] = "IsWow64Process";
8904: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
8905: GetProcAddress(module, funcName);
8906:
8907: if (NULL != fnIsWow64Process)
8908: {
8909: if (!fnIsWow64Process(GetCurrentProcess(),
8910: &bIsWow64))
8911: //throw std::exception("Unknown error");
8912: printf("Unknown error\n");
8913: }
8914: return bIsWow64 != FALSE;
8915: }
8916: #endif
1.177 brouard 8917:
1.191 brouard 8918: void syscompilerinfo(int logged)
1.167 brouard 8919: {
8920: /* #include "syscompilerinfo.h"*/
1.185 brouard 8921: /* command line Intel compiler 32bit windows, XP compatible:*/
8922: /* /GS /W3 /Gy
8923: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
8924: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
8925: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 8926: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
8927: */
8928: /* 64 bits */
1.185 brouard 8929: /*
8930: /GS /W3 /Gy
8931: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
8932: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
8933: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
8934: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
8935: /* Optimization are useless and O3 is slower than O2 */
8936: /*
8937: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
8938: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
8939: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
8940: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
8941: */
1.186 brouard 8942: /* Link is */ /* /OUT:"visual studio
1.185 brouard 8943: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
8944: /PDB:"visual studio
8945: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
8946: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
8947: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
8948: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
8949: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
8950: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
8951: uiAccess='false'"
8952: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
8953: /NOLOGO /TLBID:1
8954: */
1.177 brouard 8955: #if defined __INTEL_COMPILER
1.178 brouard 8956: #if defined(__GNUC__)
8957: struct utsname sysInfo; /* For Intel on Linux and OS/X */
8958: #endif
1.177 brouard 8959: #elif defined(__GNUC__)
1.179 brouard 8960: #ifndef __APPLE__
1.174 brouard 8961: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 8962: #endif
1.177 brouard 8963: struct utsname sysInfo;
1.178 brouard 8964: int cross = CROSS;
8965: if (cross){
8966: printf("Cross-");
1.191 brouard 8967: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 8968: }
1.174 brouard 8969: #endif
8970:
1.171 brouard 8971: #include <stdint.h>
1.178 brouard 8972:
1.191 brouard 8973: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 8974: #if defined(__clang__)
1.191 brouard 8975: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 8976: #endif
8977: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 8978: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 8979: #endif
8980: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 8981: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 8982: #endif
8983: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 8984: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 8985: #endif
8986: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 8987: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 8988: #endif
8989: #if defined(_MSC_VER)
1.191 brouard 8990: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 8991: #endif
8992: #if defined(__PGI)
1.191 brouard 8993: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 8994: #endif
8995: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 8996: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 8997: #endif
1.191 brouard 8998: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 8999:
1.167 brouard 9000: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9001: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9002: // Windows (x64 and x86)
1.191 brouard 9003: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9004: #elif __unix__ // all unices, not all compilers
9005: // Unix
1.191 brouard 9006: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9007: #elif __linux__
9008: // linux
1.191 brouard 9009: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9010: #elif __APPLE__
1.174 brouard 9011: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9012: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9013: #endif
9014:
9015: /* __MINGW32__ */
9016: /* __CYGWIN__ */
9017: /* __MINGW64__ */
9018: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9019: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9020: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9021: /* _WIN64 // Defined for applications for Win64. */
9022: /* _M_X64 // Defined for compilations that target x64 processors. */
9023: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9024:
1.167 brouard 9025: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9026: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9027: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9028: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9029: #else
1.191 brouard 9030: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9031: #endif
9032:
1.169 brouard 9033: #if defined(__GNUC__)
9034: # if defined(__GNUC_PATCHLEVEL__)
9035: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9036: + __GNUC_MINOR__ * 100 \
9037: + __GNUC_PATCHLEVEL__)
9038: # else
9039: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9040: + __GNUC_MINOR__ * 100)
9041: # endif
1.174 brouard 9042: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9043: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9044:
9045: if (uname(&sysInfo) != -1) {
9046: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9047: 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 9048: }
9049: else
9050: perror("uname() error");
1.179 brouard 9051: //#ifndef __INTEL_COMPILER
9052: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9053: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9054: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9055: #endif
1.169 brouard 9056: #endif
1.172 brouard 9057:
9058: // void main()
9059: // {
1.169 brouard 9060: #if defined(_MSC_VER)
1.174 brouard 9061: if (IsWow64()){
1.191 brouard 9062: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9063: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9064: }
9065: else{
1.191 brouard 9066: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9067: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9068: }
1.172 brouard 9069: // printf("\nPress Enter to continue...");
9070: // getchar();
9071: // }
9072:
1.169 brouard 9073: #endif
9074:
1.167 brouard 9075:
1.219 brouard 9076: }
1.136 brouard 9077:
1.219 brouard 9078: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9079: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9080: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9081: /* double ftolpl = 1.e-10; */
1.180 brouard 9082: double age, agebase, agelim;
1.203 brouard 9083: double tot;
1.180 brouard 9084:
1.202 brouard 9085: strcpy(filerespl,"PL_");
9086: strcat(filerespl,fileresu);
9087: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9088: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9089: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9090: }
1.227 brouard 9091: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9092: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9093: pstamp(ficrespl);
1.203 brouard 9094: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9095: fprintf(ficrespl,"#Age ");
9096: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9097: fprintf(ficrespl,"\n");
1.180 brouard 9098:
1.219 brouard 9099: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9100:
1.219 brouard 9101: agebase=ageminpar;
9102: agelim=agemaxpar;
1.180 brouard 9103:
1.227 brouard 9104: /* i1=pow(2,ncoveff); */
1.234 brouard 9105: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9106: if (cptcovn < 1){i1=1;}
1.180 brouard 9107:
1.238 brouard 9108: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9109: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9110: if(TKresult[nres]!= k)
9111: continue;
1.235 brouard 9112:
1.238 brouard 9113: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9114: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9115: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9116: /* k=k+1; */
9117: /* to clean */
9118: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9119: fprintf(ficrespl,"#******");
9120: printf("#******");
9121: fprintf(ficlog,"#******");
9122: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9123: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9124: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9125: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9126: }
9127: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9128: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9129: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9130: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9131: }
9132: fprintf(ficrespl,"******\n");
9133: printf("******\n");
9134: fprintf(ficlog,"******\n");
9135: if(invalidvarcomb[k]){
9136: printf("\nCombination (%d) ignored because no case \n",k);
9137: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9138: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9139: continue;
9140: }
1.219 brouard 9141:
1.238 brouard 9142: fprintf(ficrespl,"#Age ");
9143: for(j=1;j<=cptcoveff;j++) {
9144: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9145: }
9146: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9147: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9148:
1.238 brouard 9149: for (age=agebase; age<=agelim; age++){
9150: /* for (age=agebase; age<=agebase; age++){ */
9151: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9152: fprintf(ficrespl,"%.0f ",age );
9153: for(j=1;j<=cptcoveff;j++)
9154: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9155: tot=0.;
9156: for(i=1; i<=nlstate;i++){
9157: tot += prlim[i][i];
9158: fprintf(ficrespl," %.5f", prlim[i][i]);
9159: }
9160: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9161: } /* Age */
9162: /* was end of cptcod */
9163: } /* cptcov */
9164: } /* nres */
1.219 brouard 9165: return 0;
1.180 brouard 9166: }
9167:
1.218 brouard 9168: 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){
9169: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9170:
9171: /* Computes the back prevalence limit for any combination of covariate values
9172: * at any age between ageminpar and agemaxpar
9173: */
1.235 brouard 9174: int i, j, k, i1, nres=0 ;
1.217 brouard 9175: /* double ftolpl = 1.e-10; */
9176: double age, agebase, agelim;
9177: double tot;
1.218 brouard 9178: /* double ***mobaverage; */
9179: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9180:
9181: strcpy(fileresplb,"PLB_");
9182: strcat(fileresplb,fileresu);
9183: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9184: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9185: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9186: }
9187: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9188: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9189: pstamp(ficresplb);
9190: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9191: fprintf(ficresplb,"#Age ");
9192: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9193: fprintf(ficresplb,"\n");
9194:
1.218 brouard 9195:
9196: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9197:
9198: agebase=ageminpar;
9199: agelim=agemaxpar;
9200:
9201:
1.227 brouard 9202: i1=pow(2,cptcoveff);
1.218 brouard 9203: if (cptcovn < 1){i1=1;}
1.227 brouard 9204:
1.238 brouard 9205: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9206: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9207: if(TKresult[nres]!= k)
9208: continue;
9209: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9210: fprintf(ficresplb,"#******");
9211: printf("#******");
9212: fprintf(ficlog,"#******");
9213: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9214: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9215: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9216: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9217: }
9218: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9219: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9220: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9221: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9222: }
9223: fprintf(ficresplb,"******\n");
9224: printf("******\n");
9225: fprintf(ficlog,"******\n");
9226: if(invalidvarcomb[k]){
9227: printf("\nCombination (%d) ignored because no cases \n",k);
9228: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9229: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9230: continue;
9231: }
1.218 brouard 9232:
1.238 brouard 9233: fprintf(ficresplb,"#Age ");
9234: for(j=1;j<=cptcoveff;j++) {
9235: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9236: }
9237: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9238: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9239:
9240:
1.238 brouard 9241: for (age=agebase; age<=agelim; age++){
9242: /* for (age=agebase; age<=agebase; age++){ */
9243: if(mobilavproj > 0){
9244: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9245: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
9246: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k);
9247: }else if (mobilavproj == 0){
9248: 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);
9249: 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);
9250: exit(1);
9251: }else{
9252: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
9253: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k);
9254: }
9255: fprintf(ficresplb,"%.0f ",age );
9256: for(j=1;j<=cptcoveff;j++)
9257: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9258: tot=0.;
9259: for(i=1; i<=nlstate;i++){
9260: tot += bprlim[i][i];
9261: fprintf(ficresplb," %.5f", bprlim[i][i]);
9262: }
9263: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9264: } /* Age */
9265: /* was end of cptcod */
9266: } /* end of any combination */
9267: } /* end of nres */
1.218 brouard 9268: /* hBijx(p, bage, fage); */
9269: /* fclose(ficrespijb); */
9270:
9271: return 0;
1.217 brouard 9272: }
1.218 brouard 9273:
1.180 brouard 9274: int hPijx(double *p, int bage, int fage){
9275: /*------------- h Pij x at various ages ------------*/
9276:
9277: int stepsize;
9278: int agelim;
9279: int hstepm;
9280: int nhstepm;
1.235 brouard 9281: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9282:
9283: double agedeb;
9284: double ***p3mat;
9285:
1.201 brouard 9286: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9287: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9288: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9289: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9290: }
9291: printf("Computing pij: result on file '%s' \n", filerespij);
9292: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9293:
9294: stepsize=(int) (stepm+YEARM-1)/YEARM;
9295: /*if (stepm<=24) stepsize=2;*/
9296:
9297: agelim=AGESUP;
9298: hstepm=stepsize*YEARM; /* Every year of age */
9299: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9300:
1.180 brouard 9301: /* hstepm=1; aff par mois*/
9302: pstamp(ficrespij);
9303: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9304: i1= pow(2,cptcoveff);
1.218 brouard 9305: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9306: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9307: /* k=k+1; */
1.235 brouard 9308: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9309: for(k=1; k<=i1;k++){
9310: if(TKresult[nres]!= k)
9311: continue;
1.183 brouard 9312: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9313: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9314: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9315: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9316: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9317: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9318: }
1.183 brouard 9319: fprintf(ficrespij,"******\n");
9320:
9321: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9322: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9323: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9324:
9325: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9326:
1.183 brouard 9327: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9328: oldm=oldms;savm=savms;
1.235 brouard 9329: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9330: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9331: for(i=1; i<=nlstate;i++)
9332: for(j=1; j<=nlstate+ndeath;j++)
9333: fprintf(ficrespij," %1d-%1d",i,j);
9334: fprintf(ficrespij,"\n");
9335: for (h=0; h<=nhstepm; h++){
9336: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9337: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9338: for(i=1; i<=nlstate;i++)
9339: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9340: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9341: fprintf(ficrespij,"\n");
9342: }
1.183 brouard 9343: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9344: fprintf(ficrespij,"\n");
9345: }
1.180 brouard 9346: /*}*/
9347: }
1.218 brouard 9348: return 0;
1.180 brouard 9349: }
1.218 brouard 9350:
9351: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9352: /*------------- h Bij x at various ages ------------*/
9353:
9354: int stepsize;
1.218 brouard 9355: /* int agelim; */
9356: int ageminl;
1.217 brouard 9357: int hstepm;
9358: int nhstepm;
1.238 brouard 9359: int h, i, i1, j, k, nres;
1.218 brouard 9360:
1.217 brouard 9361: double agedeb;
9362: double ***p3mat;
1.218 brouard 9363:
9364: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9365: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9366: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9367: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9368: }
9369: printf("Computing pij back: result on file '%s' \n", filerespijb);
9370: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9371:
9372: stepsize=(int) (stepm+YEARM-1)/YEARM;
9373: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9374:
1.218 brouard 9375: /* agelim=AGESUP; */
9376: ageminl=30;
9377: hstepm=stepsize*YEARM; /* Every year of age */
9378: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9379:
9380: /* hstepm=1; aff par mois*/
9381: pstamp(ficrespijb);
9382: fprintf(ficrespijb,"#****** h Pij x Back Probability to be in state i at age x-h being in j at x ");
1.227 brouard 9383: i1= pow(2,cptcoveff);
1.218 brouard 9384: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9385: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9386: /* k=k+1; */
1.238 brouard 9387: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9388: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9389: if(TKresult[nres]!= k)
9390: continue;
9391: fprintf(ficrespijb,"\n#****** ");
9392: for(j=1;j<=cptcoveff;j++)
9393: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9394: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9395: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9396: }
9397: fprintf(ficrespijb,"******\n");
9398: if(invalidvarcomb[k]){
9399: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9400: continue;
9401: }
9402:
9403: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9404: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9405: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9406: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9407: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9408:
9409: /* nhstepm=nhstepm*YEARM; aff par mois*/
9410:
9411: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9412: /* oldm=oldms;savm=savms; */
9413: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9414: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9415: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
9416: fprintf(ficrespijb,"# Cov Agex agex-h hpijx with i,j=");
1.217 brouard 9417: for(i=1; i<=nlstate;i++)
9418: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9419: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9420: fprintf(ficrespijb,"\n");
1.238 brouard 9421: for (h=0; h<=nhstepm; h++){
9422: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9423: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9424: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9425: for(i=1; i<=nlstate;i++)
9426: for(j=1; j<=nlstate+ndeath;j++)
9427: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9428: fprintf(ficrespijb,"\n");
9429: }
9430: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9431: fprintf(ficrespijb,"\n");
9432: } /* end age deb */
9433: } /* end combination */
9434: } /* end nres */
1.218 brouard 9435: return 0;
9436: } /* hBijx */
1.217 brouard 9437:
1.180 brouard 9438:
1.136 brouard 9439: /***********************************************/
9440: /**************** Main Program *****************/
9441: /***********************************************/
9442:
9443: int main(int argc, char *argv[])
9444: {
9445: #ifdef GSL
9446: const gsl_multimin_fminimizer_type *T;
9447: size_t iteri = 0, it;
9448: int rval = GSL_CONTINUE;
9449: int status = GSL_SUCCESS;
9450: double ssval;
9451: #endif
9452: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9453: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9454: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9455: int jj, ll, li, lj, lk;
1.136 brouard 9456: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9457: int num_filled;
1.136 brouard 9458: int itimes;
9459: int NDIM=2;
9460: int vpopbased=0;
1.235 brouard 9461: int nres=0;
1.136 brouard 9462:
1.164 brouard 9463: char ca[32], cb[32];
1.136 brouard 9464: /* FILE *fichtm; *//* Html File */
9465: /* FILE *ficgp;*/ /*Gnuplot File */
9466: struct stat info;
1.191 brouard 9467: double agedeb=0.;
1.194 brouard 9468:
9469: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9470: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9471:
1.165 brouard 9472: double fret;
1.191 brouard 9473: double dum=0.; /* Dummy variable */
1.136 brouard 9474: double ***p3mat;
1.218 brouard 9475: /* double ***mobaverage; */
1.164 brouard 9476:
9477: char line[MAXLINE];
1.197 brouard 9478: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9479:
1.234 brouard 9480: char modeltemp[MAXLINE];
1.230 brouard 9481: char resultline[MAXLINE];
9482:
1.136 brouard 9483: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9484: char *tok, *val; /* pathtot */
1.136 brouard 9485: int firstobs=1, lastobs=10;
1.195 brouard 9486: int c, h , cpt, c2;
1.191 brouard 9487: int jl=0;
9488: int i1, j1, jk, stepsize=0;
1.194 brouard 9489: int count=0;
9490:
1.164 brouard 9491: int *tab;
1.136 brouard 9492: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9493: int backcast=0;
1.136 brouard 9494: int mobilav=0,popforecast=0;
1.191 brouard 9495: int hstepm=0, nhstepm=0;
1.136 brouard 9496: int agemortsup;
9497: float sumlpop=0.;
9498: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9499: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9500:
1.191 brouard 9501: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9502: double ftolpl=FTOL;
9503: double **prlim;
1.217 brouard 9504: double **bprlim;
1.136 brouard 9505: double ***param; /* Matrix of parameters */
9506: double *p;
9507: double **matcov; /* Matrix of covariance */
1.203 brouard 9508: double **hess; /* Hessian matrix */
1.136 brouard 9509: double ***delti3; /* Scale */
9510: double *delti; /* Scale */
9511: double ***eij, ***vareij;
9512: double **varpl; /* Variances of prevalence limits by age */
9513: double *epj, vepp;
1.164 brouard 9514:
1.136 brouard 9515: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9516: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9517:
1.136 brouard 9518: double **ximort;
1.145 brouard 9519: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9520: int *dcwave;
9521:
1.164 brouard 9522: char z[1]="c";
1.136 brouard 9523:
9524: /*char *strt;*/
9525: char strtend[80];
1.126 brouard 9526:
1.164 brouard 9527:
1.126 brouard 9528: /* setlocale (LC_ALL, ""); */
9529: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9530: /* textdomain (PACKAGE); */
9531: /* setlocale (LC_CTYPE, ""); */
9532: /* setlocale (LC_MESSAGES, ""); */
9533:
9534: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9535: rstart_time = time(NULL);
9536: /* (void) gettimeofday(&start_time,&tzp);*/
9537: start_time = *localtime(&rstart_time);
1.126 brouard 9538: curr_time=start_time;
1.157 brouard 9539: /*tml = *localtime(&start_time.tm_sec);*/
9540: /* strcpy(strstart,asctime(&tml)); */
9541: strcpy(strstart,asctime(&start_time));
1.126 brouard 9542:
9543: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9544: /* tp.tm_sec = tp.tm_sec +86400; */
9545: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9546: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9547: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9548: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9549: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9550: /* strt=asctime(&tmg); */
9551: /* printf("Time(after) =%s",strstart); */
9552: /* (void) time (&time_value);
9553: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9554: * tm = *localtime(&time_value);
9555: * strstart=asctime(&tm);
9556: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9557: */
9558:
9559: nberr=0; /* Number of errors and warnings */
9560: nbwarn=0;
1.184 brouard 9561: #ifdef WIN32
9562: _getcwd(pathcd, size);
9563: #else
1.126 brouard 9564: getcwd(pathcd, size);
1.184 brouard 9565: #endif
1.191 brouard 9566: syscompilerinfo(0);
1.196 brouard 9567: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9568: if(argc <=1){
9569: printf("\nEnter the parameter file name: ");
1.205 brouard 9570: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9571: printf("ERROR Empty parameter file name\n");
9572: goto end;
9573: }
1.126 brouard 9574: i=strlen(pathr);
9575: if(pathr[i-1]=='\n')
9576: pathr[i-1]='\0';
1.156 brouard 9577: i=strlen(pathr);
1.205 brouard 9578: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9579: pathr[i-1]='\0';
1.205 brouard 9580: }
9581: i=strlen(pathr);
9582: if( i==0 ){
9583: printf("ERROR Empty parameter file name\n");
9584: goto end;
9585: }
9586: for (tok = pathr; tok != NULL; ){
1.126 brouard 9587: printf("Pathr |%s|\n",pathr);
9588: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9589: printf("val= |%s| pathr=%s\n",val,pathr);
9590: strcpy (pathtot, val);
9591: if(pathr[0] == '\0') break; /* Dirty */
9592: }
9593: }
9594: else{
9595: strcpy(pathtot,argv[1]);
9596: }
9597: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9598: /*cygwin_split_path(pathtot,path,optionfile);
9599: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9600: /* cutv(path,optionfile,pathtot,'\\');*/
9601:
9602: /* Split argv[0], imach program to get pathimach */
9603: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9604: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9605: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9606: /* strcpy(pathimach,argv[0]); */
9607: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9608: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9609: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9610: #ifdef WIN32
9611: _chdir(path); /* Can be a relative path */
9612: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9613: #else
1.126 brouard 9614: chdir(path); /* Can be a relative path */
1.184 brouard 9615: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9616: #endif
9617: printf("Current directory %s!\n",pathcd);
1.126 brouard 9618: strcpy(command,"mkdir ");
9619: strcat(command,optionfilefiname);
9620: if((outcmd=system(command)) != 0){
1.169 brouard 9621: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9622: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9623: /* fclose(ficlog); */
9624: /* exit(1); */
9625: }
9626: /* if((imk=mkdir(optionfilefiname))<0){ */
9627: /* perror("mkdir"); */
9628: /* } */
9629:
9630: /*-------- arguments in the command line --------*/
9631:
1.186 brouard 9632: /* Main Log file */
1.126 brouard 9633: strcat(filelog, optionfilefiname);
9634: strcat(filelog,".log"); /* */
9635: if((ficlog=fopen(filelog,"w"))==NULL) {
9636: printf("Problem with logfile %s\n",filelog);
9637: goto end;
9638: }
9639: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9640: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9641: fprintf(ficlog,"\nEnter the parameter file name: \n");
9642: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9643: path=%s \n\
9644: optionfile=%s\n\
9645: optionfilext=%s\n\
1.156 brouard 9646: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9647:
1.197 brouard 9648: syscompilerinfo(1);
1.167 brouard 9649:
1.126 brouard 9650: printf("Local time (at start):%s",strstart);
9651: fprintf(ficlog,"Local time (at start): %s",strstart);
9652: fflush(ficlog);
9653: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9654: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9655:
9656: /* */
9657: strcpy(fileres,"r");
9658: strcat(fileres, optionfilefiname);
1.201 brouard 9659: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9660: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9661: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9662:
1.186 brouard 9663: /* Main ---------arguments file --------*/
1.126 brouard 9664:
9665: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9666: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9667: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9668: fflush(ficlog);
1.149 brouard 9669: /* goto end; */
9670: exit(70);
1.126 brouard 9671: }
9672:
9673:
9674:
9675: strcpy(filereso,"o");
1.201 brouard 9676: strcat(filereso,fileresu);
1.126 brouard 9677: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9678: printf("Problem with Output resultfile: %s\n", filereso);
9679: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9680: fflush(ficlog);
9681: goto end;
9682: }
9683:
9684: /* Reads comments: lines beginning with '#' */
9685: numlinepar=0;
1.197 brouard 9686:
9687: /* First parameter line */
9688: while(fgets(line, MAXLINE, ficpar)) {
9689: /* If line starts with a # it is a comment */
9690: if (line[0] == '#') {
9691: numlinepar++;
9692: fputs(line,stdout);
9693: fputs(line,ficparo);
9694: fputs(line,ficlog);
9695: continue;
9696: }else
9697: break;
9698: }
9699: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
9700: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
9701: if (num_filled != 5) {
9702: printf("Should be 5 parameters\n");
9703: }
1.126 brouard 9704: numlinepar++;
1.197 brouard 9705: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
9706: }
9707: /* Second parameter line */
9708: while(fgets(line, MAXLINE, ficpar)) {
9709: /* If line starts with a # it is a comment */
9710: if (line[0] == '#') {
9711: numlinepar++;
9712: fputs(line,stdout);
9713: fputs(line,ficparo);
9714: fputs(line,ficlog);
9715: continue;
9716: }else
9717: break;
9718: }
1.223 brouard 9719: 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", \
9720: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
9721: if (num_filled != 11) {
9722: 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 9723: printf("but line=%s\n",line);
1.197 brouard 9724: }
1.223 brouard 9725: 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 9726: }
1.203 brouard 9727: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 9728: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 9729: /* Third parameter line */
9730: while(fgets(line, MAXLINE, ficpar)) {
9731: /* If line starts with a # it is a comment */
9732: if (line[0] == '#') {
9733: numlinepar++;
9734: fputs(line,stdout);
9735: fputs(line,ficparo);
9736: fputs(line,ficlog);
9737: continue;
9738: }else
9739: break;
9740: }
1.201 brouard 9741: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
9742: if (num_filled == 0)
9743: model[0]='\0';
9744: else if (num_filled != 1){
1.197 brouard 9745: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9746: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9747: model[0]='\0';
9748: goto end;
9749: }
9750: else{
9751: if (model[0]=='+'){
9752: for(i=1; i<=strlen(model);i++)
9753: modeltemp[i-1]=model[i];
1.201 brouard 9754: strcpy(model,modeltemp);
1.197 brouard 9755: }
9756: }
1.199 brouard 9757: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 9758: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 9759: }
9760: /* 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); */
9761: /* numlinepar=numlinepar+3; /\* In general *\/ */
9762: /* 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 9763: 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);
9764: 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 9765: fflush(ficlog);
1.190 brouard 9766: /* if(model[0]=='#'|| model[0]== '\0'){ */
9767: if(model[0]=='#'){
1.187 brouard 9768: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
9769: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
9770: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
9771: if(mle != -1){
9772: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
9773: exit(1);
9774: }
9775: }
1.126 brouard 9776: while((c=getc(ficpar))=='#' && c!= EOF){
9777: ungetc(c,ficpar);
9778: fgets(line, MAXLINE, ficpar);
9779: numlinepar++;
1.195 brouard 9780: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
9781: z[0]=line[1];
9782: }
9783: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 9784: fputs(line, stdout);
9785: //puts(line);
1.126 brouard 9786: fputs(line,ficparo);
9787: fputs(line,ficlog);
9788: }
9789: ungetc(c,ficpar);
9790:
9791:
1.145 brouard 9792: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 9793: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 9794: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 9795: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 9796: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
9797: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
9798: v1+v2*age+v2*v3 makes cptcovn = 3
9799: */
9800: if (strlen(model)>1)
1.187 brouard 9801: 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 9802: else
1.187 brouard 9803: ncovmodel=2; /* Constant and age */
1.133 brouard 9804: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
9805: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 9806: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
9807: 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);
9808: 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);
9809: fflush(stdout);
9810: fclose (ficlog);
9811: goto end;
9812: }
1.126 brouard 9813: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9814: delti=delti3[1][1];
9815: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
9816: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
9817: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 9818: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
9819: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9820: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
9821: fclose (ficparo);
9822: fclose (ficlog);
9823: goto end;
9824: exit(0);
1.220 brouard 9825: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 9826: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 9827: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
9828: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9829: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9830: matcov=matrix(1,npar,1,npar);
1.203 brouard 9831: hess=matrix(1,npar,1,npar);
1.220 brouard 9832: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 9833: /* Read guessed parameters */
1.126 brouard 9834: /* Reads comments: lines beginning with '#' */
9835: while((c=getc(ficpar))=='#' && c!= EOF){
9836: ungetc(c,ficpar);
9837: fgets(line, MAXLINE, ficpar);
9838: numlinepar++;
1.141 brouard 9839: fputs(line,stdout);
1.126 brouard 9840: fputs(line,ficparo);
9841: fputs(line,ficlog);
9842: }
9843: ungetc(c,ficpar);
9844:
9845: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9846: for(i=1; i <=nlstate; i++){
1.234 brouard 9847: j=0;
1.126 brouard 9848: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 9849: if(jj==i) continue;
9850: j++;
9851: fscanf(ficpar,"%1d%1d",&i1,&j1);
9852: if ((i1 != i) || (j1 != jj)){
9853: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 9854: It might be a problem of design; if ncovcol and the model are correct\n \
9855: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 9856: exit(1);
9857: }
9858: fprintf(ficparo,"%1d%1d",i1,j1);
9859: if(mle==1)
9860: printf("%1d%1d",i,jj);
9861: fprintf(ficlog,"%1d%1d",i,jj);
9862: for(k=1; k<=ncovmodel;k++){
9863: fscanf(ficpar," %lf",¶m[i][j][k]);
9864: if(mle==1){
9865: printf(" %lf",param[i][j][k]);
9866: fprintf(ficlog," %lf",param[i][j][k]);
9867: }
9868: else
9869: fprintf(ficlog," %lf",param[i][j][k]);
9870: fprintf(ficparo," %lf",param[i][j][k]);
9871: }
9872: fscanf(ficpar,"\n");
9873: numlinepar++;
9874: if(mle==1)
9875: printf("\n");
9876: fprintf(ficlog,"\n");
9877: fprintf(ficparo,"\n");
1.126 brouard 9878: }
9879: }
9880: fflush(ficlog);
1.234 brouard 9881:
1.145 brouard 9882: /* Reads scales values */
1.126 brouard 9883: p=param[1][1];
9884:
9885: /* Reads comments: lines beginning with '#' */
9886: while((c=getc(ficpar))=='#' && c!= EOF){
9887: ungetc(c,ficpar);
9888: fgets(line, MAXLINE, ficpar);
9889: numlinepar++;
1.141 brouard 9890: fputs(line,stdout);
1.126 brouard 9891: fputs(line,ficparo);
9892: fputs(line,ficlog);
9893: }
9894: ungetc(c,ficpar);
9895:
9896: for(i=1; i <=nlstate; i++){
9897: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 9898: fscanf(ficpar,"%1d%1d",&i1,&j1);
9899: if ( (i1-i) * (j1-j) != 0){
9900: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
9901: exit(1);
9902: }
9903: printf("%1d%1d",i,j);
9904: fprintf(ficparo,"%1d%1d",i1,j1);
9905: fprintf(ficlog,"%1d%1d",i1,j1);
9906: for(k=1; k<=ncovmodel;k++){
9907: fscanf(ficpar,"%le",&delti3[i][j][k]);
9908: printf(" %le",delti3[i][j][k]);
9909: fprintf(ficparo," %le",delti3[i][j][k]);
9910: fprintf(ficlog," %le",delti3[i][j][k]);
9911: }
9912: fscanf(ficpar,"\n");
9913: numlinepar++;
9914: printf("\n");
9915: fprintf(ficparo,"\n");
9916: fprintf(ficlog,"\n");
1.126 brouard 9917: }
9918: }
9919: fflush(ficlog);
1.234 brouard 9920:
1.145 brouard 9921: /* Reads covariance matrix */
1.126 brouard 9922: delti=delti3[1][1];
1.220 brouard 9923:
9924:
1.126 brouard 9925: /* 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 9926:
1.126 brouard 9927: /* Reads comments: lines beginning with '#' */
9928: while((c=getc(ficpar))=='#' && c!= EOF){
9929: ungetc(c,ficpar);
9930: fgets(line, MAXLINE, ficpar);
9931: numlinepar++;
1.141 brouard 9932: fputs(line,stdout);
1.126 brouard 9933: fputs(line,ficparo);
9934: fputs(line,ficlog);
9935: }
9936: ungetc(c,ficpar);
1.220 brouard 9937:
1.126 brouard 9938: matcov=matrix(1,npar,1,npar);
1.203 brouard 9939: hess=matrix(1,npar,1,npar);
1.131 brouard 9940: for(i=1; i <=npar; i++)
9941: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 9942:
1.194 brouard 9943: /* Scans npar lines */
1.126 brouard 9944: for(i=1; i <=npar; i++){
1.226 brouard 9945: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 9946: if(count != 3){
1.226 brouard 9947: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 9948: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
9949: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 9950: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 9951: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
9952: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 9953: exit(1);
1.220 brouard 9954: }else{
1.226 brouard 9955: if(mle==1)
9956: printf("%1d%1d%d",i1,j1,jk);
9957: }
9958: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
9959: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 9960: for(j=1; j <=i; j++){
1.226 brouard 9961: fscanf(ficpar," %le",&matcov[i][j]);
9962: if(mle==1){
9963: printf(" %.5le",matcov[i][j]);
9964: }
9965: fprintf(ficlog," %.5le",matcov[i][j]);
9966: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 9967: }
9968: fscanf(ficpar,"\n");
9969: numlinepar++;
9970: if(mle==1)
1.220 brouard 9971: printf("\n");
1.126 brouard 9972: fprintf(ficlog,"\n");
9973: fprintf(ficparo,"\n");
9974: }
1.194 brouard 9975: /* End of read covariance matrix npar lines */
1.126 brouard 9976: for(i=1; i <=npar; i++)
9977: for(j=i+1;j<=npar;j++)
1.226 brouard 9978: matcov[i][j]=matcov[j][i];
1.126 brouard 9979:
9980: if(mle==1)
9981: printf("\n");
9982: fprintf(ficlog,"\n");
9983:
9984: fflush(ficlog);
9985:
9986: /*-------- Rewriting parameter file ----------*/
9987: strcpy(rfileres,"r"); /* "Rparameterfile */
9988: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
9989: strcat(rfileres,"."); /* */
9990: strcat(rfileres,optionfilext); /* Other files have txt extension */
9991: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 9992: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
9993: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 9994: }
9995: fprintf(ficres,"#%s\n",version);
9996: } /* End of mle != -3 */
1.218 brouard 9997:
1.186 brouard 9998: /* Main data
9999: */
1.126 brouard 10000: n= lastobs;
10001: num=lvector(1,n);
10002: moisnais=vector(1,n);
10003: annais=vector(1,n);
10004: moisdc=vector(1,n);
10005: andc=vector(1,n);
1.220 brouard 10006: weight=vector(1,n);
1.126 brouard 10007: agedc=vector(1,n);
10008: cod=ivector(1,n);
1.220 brouard 10009: for(i=1;i<=n;i++){
1.234 brouard 10010: num[i]=0;
10011: moisnais[i]=0;
10012: annais[i]=0;
10013: moisdc[i]=0;
10014: andc[i]=0;
10015: agedc[i]=0;
10016: cod[i]=0;
10017: weight[i]=1.0; /* Equal weights, 1 by default */
10018: }
1.126 brouard 10019: mint=matrix(1,maxwav,1,n);
10020: anint=matrix(1,maxwav,1,n);
1.131 brouard 10021: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10022: tab=ivector(1,NCOVMAX);
1.144 brouard 10023: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10024: 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 10025:
1.136 brouard 10026: /* Reads data from file datafile */
10027: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10028: goto end;
10029:
10030: /* Calculation of the number of parameters from char model */
1.234 brouard 10031: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10032: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10033: k=3 V4 Tvar[k=3]= 4 (from V4)
10034: k=2 V1 Tvar[k=2]= 1 (from V1)
10035: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10036: */
10037:
10038: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10039: TvarsDind=ivector(1,NCOVMAX); /* */
10040: TvarsD=ivector(1,NCOVMAX); /* */
10041: TvarsQind=ivector(1,NCOVMAX); /* */
10042: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10043: TvarF=ivector(1,NCOVMAX); /* */
10044: TvarFind=ivector(1,NCOVMAX); /* */
10045: TvarV=ivector(1,NCOVMAX); /* */
10046: TvarVind=ivector(1,NCOVMAX); /* */
10047: TvarA=ivector(1,NCOVMAX); /* */
10048: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10049: TvarFD=ivector(1,NCOVMAX); /* */
10050: TvarFDind=ivector(1,NCOVMAX); /* */
10051: TvarFQ=ivector(1,NCOVMAX); /* */
10052: TvarFQind=ivector(1,NCOVMAX); /* */
10053: TvarVD=ivector(1,NCOVMAX); /* */
10054: TvarVDind=ivector(1,NCOVMAX); /* */
10055: TvarVQ=ivector(1,NCOVMAX); /* */
10056: TvarVQind=ivector(1,NCOVMAX); /* */
10057:
1.230 brouard 10058: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10059: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10060: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10061: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10062: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10063: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10064: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10065: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10066: */
10067: /* For model-covariate k tells which data-covariate to use but
10068: because this model-covariate is a construction we invent a new column
10069: ncovcol + k1
10070: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10071: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10072: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10073: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10074: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10075: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10076: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10077: */
1.145 brouard 10078: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10079: 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 10080: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10081: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10082: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10083: 4 covariates (3 plus signs)
10084: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10085: */
1.230 brouard 10086: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10087: * individual dummy, fixed or varying:
10088: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10089: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10090: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10091: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10092: * Tmodelind[1]@9={9,0,3,2,}*/
10093: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10094: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10095: * individual quantitative, fixed or varying:
10096: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10097: * 3, 1, 0, 0, 0, 0, 0, 0},
10098: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10099: /* Main decodemodel */
10100:
1.187 brouard 10101:
1.223 brouard 10102: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10103: goto end;
10104:
1.137 brouard 10105: if((double)(lastobs-imx)/(double)imx > 1.10){
10106: nbwarn++;
10107: 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);
10108: 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);
10109: }
1.136 brouard 10110: /* if(mle==1){*/
1.137 brouard 10111: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10112: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10113: }
10114:
10115: /*-calculation of age at interview from date of interview and age at death -*/
10116: agev=matrix(1,maxwav,1,imx);
10117:
10118: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10119: goto end;
10120:
1.126 brouard 10121:
1.136 brouard 10122: agegomp=(int)agemin;
10123: free_vector(moisnais,1,n);
10124: free_vector(annais,1,n);
1.126 brouard 10125: /* free_matrix(mint,1,maxwav,1,n);
10126: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10127: /* free_vector(moisdc,1,n); */
10128: /* free_vector(andc,1,n); */
1.145 brouard 10129: /* */
10130:
1.126 brouard 10131: wav=ivector(1,imx);
1.214 brouard 10132: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10133: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10134: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10135: 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.*/
10136: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10137: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10138:
10139: /* Concatenates waves */
1.214 brouard 10140: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10141: Death is a valid wave (if date is known).
10142: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10143: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10144: and mw[mi+1][i]. dh depends on stepm.
10145: */
10146:
1.126 brouard 10147: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.145 brouard 10148: /* */
10149:
1.215 brouard 10150: free_vector(moisdc,1,n);
10151: free_vector(andc,1,n);
10152:
1.126 brouard 10153: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10154: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10155: ncodemax[1]=1;
1.145 brouard 10156: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10157: cptcoveff=0;
1.220 brouard 10158: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10159: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10160: }
10161:
10162: ncovcombmax=pow(2,cptcoveff);
10163: invalidvarcomb=ivector(1, ncovcombmax);
10164: for(i=1;i<ncovcombmax;i++)
10165: invalidvarcomb[i]=0;
10166:
1.211 brouard 10167: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10168: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10169: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10170:
1.200 brouard 10171: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10172: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10173: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10174: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10175: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10176: * (currently 0 or 1) in the data.
10177: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10178: * corresponding modality (h,j).
10179: */
10180:
1.145 brouard 10181: h=0;
10182: /*if (cptcovn > 0) */
1.126 brouard 10183: m=pow(2,cptcoveff);
10184:
1.144 brouard 10185: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10186: * For k=4 covariates, h goes from 1 to m=2**k
10187: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10188: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10189: * h\k 1 2 3 4
1.143 brouard 10190: *______________________________
10191: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10192: * 2 2 1 1 1
10193: * 3 i=2 1 2 1 1
10194: * 4 2 2 1 1
10195: * 5 i=3 1 i=2 1 2 1
10196: * 6 2 1 2 1
10197: * 7 i=4 1 2 2 1
10198: * 8 2 2 2 1
1.197 brouard 10199: * 9 i=5 1 i=3 1 i=2 1 2
10200: * 10 2 1 1 2
10201: * 11 i=6 1 2 1 2
10202: * 12 2 2 1 2
10203: * 13 i=7 1 i=4 1 2 2
10204: * 14 2 1 2 2
10205: * 15 i=8 1 2 2 2
10206: * 16 2 2 2 2
1.143 brouard 10207: */
1.212 brouard 10208: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10209: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10210: * and the value of each covariate?
10211: * V1=1, V2=1, V3=2, V4=1 ?
10212: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10213: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10214: * In order to get the real value in the data, we use nbcode
10215: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10216: * We are keeping this crazy system in order to be able (in the future?)
10217: * to have more than 2 values (0 or 1) for a covariate.
10218: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10219: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10220: * bbbbbbbb
10221: * 76543210
10222: * h-1 00000101 (6-1=5)
1.219 brouard 10223: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10224: * &
10225: * 1 00000001 (1)
1.219 brouard 10226: * 00000000 = 1 & ((h-1) >> (k-1))
10227: * +1= 00000001 =1
1.211 brouard 10228: *
10229: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10230: * h' 1101 =2^3+2^2+0x2^1+2^0
10231: * >>k' 11
10232: * & 00000001
10233: * = 00000001
10234: * +1 = 00000010=2 = codtabm(14,3)
10235: * Reverse h=6 and m=16?
10236: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10237: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10238: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10239: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10240: * V3=decodtabm(14,3,2**4)=2
10241: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10242: *(h-1) >> (j-1) 0011 =13 >> 2
10243: * &1 000000001
10244: * = 000000001
10245: * +1= 000000010 =2
10246: * 2211
10247: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10248: * V3=2
1.220 brouard 10249: * codtabm and decodtabm are identical
1.211 brouard 10250: */
10251:
1.145 brouard 10252:
10253: free_ivector(Ndum,-1,NCOVMAX);
10254:
10255:
1.126 brouard 10256:
1.186 brouard 10257: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10258: strcpy(optionfilegnuplot,optionfilefiname);
10259: if(mle==-3)
1.201 brouard 10260: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10261: strcat(optionfilegnuplot,".gp");
10262:
10263: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10264: printf("Problem with file %s",optionfilegnuplot);
10265: }
10266: else{
1.204 brouard 10267: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10268: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10269: //fprintf(ficgp,"set missing 'NaNq'\n");
10270: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10271: }
10272: /* fclose(ficgp);*/
1.186 brouard 10273:
10274:
10275: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10276:
10277: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10278: if(mle==-3)
1.201 brouard 10279: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10280: strcat(optionfilehtm,".htm");
10281: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10282: printf("Problem with %s \n",optionfilehtm);
10283: exit(0);
1.126 brouard 10284: }
10285:
10286: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10287: strcat(optionfilehtmcov,"-cov.htm");
10288: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10289: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10290: }
10291: else{
10292: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10293: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10294: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10295: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10296: }
10297:
1.213 brouard 10298: 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 10299: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10300: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10301: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10302: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10303: \n\
10304: <hr size=\"2\" color=\"#EC5E5E\">\
10305: <ul><li><h4>Parameter files</h4>\n\
10306: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10307: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10308: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10309: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10310: - Date and time at start: %s</ul>\n",\
10311: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10312: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10313: fileres,fileres,\
10314: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10315: fflush(fichtm);
10316:
10317: strcpy(pathr,path);
10318: strcat(pathr,optionfilefiname);
1.184 brouard 10319: #ifdef WIN32
10320: _chdir(optionfilefiname); /* Move to directory named optionfile */
10321: #else
1.126 brouard 10322: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10323: #endif
10324:
1.126 brouard 10325:
1.220 brouard 10326: /* Calculates basic frequencies. Computes observed prevalence at single age
10327: and for any valid combination of covariates
1.126 brouard 10328: and prints on file fileres'p'. */
1.227 brouard 10329: freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
10330: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10331:
10332: fprintf(fichtm,"\n");
10333: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10334: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10335: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10336: imx,agemin,agemax,jmin,jmax,jmean);
10337: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10338: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10339: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10340: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10341: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10342:
1.126 brouard 10343: /* For Powell, parameters are in a vector p[] starting at p[1]
10344: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10345: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10346:
10347: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10348: /* For mortality only */
1.126 brouard 10349: if (mle==-3){
1.136 brouard 10350: ximort=matrix(1,NDIM,1,NDIM);
1.220 brouard 10351: for(i=1;i<=NDIM;i++)
10352: for(j=1;j<=NDIM;j++)
10353: ximort[i][j]=0.;
1.186 brouard 10354: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10355: cens=ivector(1,n);
10356: ageexmed=vector(1,n);
10357: agecens=vector(1,n);
10358: dcwave=ivector(1,n);
1.223 brouard 10359:
1.126 brouard 10360: for (i=1; i<=imx; i++){
10361: dcwave[i]=-1;
10362: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10363: if (s[m][i]>nlstate) {
10364: dcwave[i]=m;
10365: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10366: break;
10367: }
1.126 brouard 10368: }
1.226 brouard 10369:
1.126 brouard 10370: for (i=1; i<=imx; i++) {
10371: if (wav[i]>0){
1.226 brouard 10372: ageexmed[i]=agev[mw[1][i]][i];
10373: j=wav[i];
10374: agecens[i]=1.;
10375:
10376: if (ageexmed[i]> 1 && wav[i] > 0){
10377: agecens[i]=agev[mw[j][i]][i];
10378: cens[i]= 1;
10379: }else if (ageexmed[i]< 1)
10380: cens[i]= -1;
10381: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10382: cens[i]=0 ;
1.126 brouard 10383: }
10384: else cens[i]=-1;
10385: }
10386:
10387: for (i=1;i<=NDIM;i++) {
10388: for (j=1;j<=NDIM;j++)
1.226 brouard 10389: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10390: }
10391:
1.145 brouard 10392: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10393: /*printf("%lf %lf", p[1], p[2]);*/
10394:
10395:
1.136 brouard 10396: #ifdef GSL
10397: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10398: #else
1.126 brouard 10399: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10400: #endif
1.201 brouard 10401: strcpy(filerespow,"POW-MORT_");
10402: strcat(filerespow,fileresu);
1.126 brouard 10403: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10404: printf("Problem with resultfile: %s\n", filerespow);
10405: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10406: }
1.136 brouard 10407: #ifdef GSL
10408: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10409: #else
1.126 brouard 10410: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10411: #endif
1.126 brouard 10412: /* for (i=1;i<=nlstate;i++)
10413: for(j=1;j<=nlstate+ndeath;j++)
10414: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10415: */
10416: fprintf(ficrespow,"\n");
1.136 brouard 10417: #ifdef GSL
10418: /* gsl starts here */
10419: T = gsl_multimin_fminimizer_nmsimplex;
10420: gsl_multimin_fminimizer *sfm = NULL;
10421: gsl_vector *ss, *x;
10422: gsl_multimin_function minex_func;
10423:
10424: /* Initial vertex size vector */
10425: ss = gsl_vector_alloc (NDIM);
10426:
10427: if (ss == NULL){
10428: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10429: }
10430: /* Set all step sizes to 1 */
10431: gsl_vector_set_all (ss, 0.001);
10432:
10433: /* Starting point */
1.126 brouard 10434:
1.136 brouard 10435: x = gsl_vector_alloc (NDIM);
10436:
10437: if (x == NULL){
10438: gsl_vector_free(ss);
10439: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10440: }
10441:
10442: /* Initialize method and iterate */
10443: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10444: /* gsl_vector_set(x, 0, 0.0268); */
10445: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10446: gsl_vector_set(x, 0, p[1]);
10447: gsl_vector_set(x, 1, p[2]);
10448:
10449: minex_func.f = &gompertz_f;
10450: minex_func.n = NDIM;
10451: minex_func.params = (void *)&p; /* ??? */
10452:
10453: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10454: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10455:
10456: printf("Iterations beginning .....\n\n");
10457: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10458:
10459: iteri=0;
10460: while (rval == GSL_CONTINUE){
10461: iteri++;
10462: status = gsl_multimin_fminimizer_iterate(sfm);
10463:
10464: if (status) printf("error: %s\n", gsl_strerror (status));
10465: fflush(0);
10466:
10467: if (status)
10468: break;
10469:
10470: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10471: ssval = gsl_multimin_fminimizer_size (sfm);
10472:
10473: if (rval == GSL_SUCCESS)
10474: printf ("converged to a local maximum at\n");
10475:
10476: printf("%5d ", iteri);
10477: for (it = 0; it < NDIM; it++){
10478: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10479: }
10480: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10481: }
10482:
10483: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10484:
10485: gsl_vector_free(x); /* initial values */
10486: gsl_vector_free(ss); /* inital step size */
10487: for (it=0; it<NDIM; it++){
10488: p[it+1]=gsl_vector_get(sfm->x,it);
10489: fprintf(ficrespow," %.12lf", p[it]);
10490: }
10491: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10492: #endif
10493: #ifdef POWELL
10494: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10495: #endif
1.126 brouard 10496: fclose(ficrespow);
10497:
1.203 brouard 10498: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10499:
10500: for(i=1; i <=NDIM; i++)
10501: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10502: matcov[i][j]=matcov[j][i];
1.126 brouard 10503:
10504: printf("\nCovariance matrix\n ");
1.203 brouard 10505: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10506: for(i=1; i <=NDIM; i++) {
10507: for(j=1;j<=NDIM;j++){
1.220 brouard 10508: printf("%f ",matcov[i][j]);
10509: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10510: }
1.203 brouard 10511: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10512: }
10513:
10514: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10515: for (i=1;i<=NDIM;i++) {
1.126 brouard 10516: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10517: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10518: }
1.126 brouard 10519: lsurv=vector(1,AGESUP);
10520: lpop=vector(1,AGESUP);
10521: tpop=vector(1,AGESUP);
10522: lsurv[agegomp]=100000;
10523:
10524: for (k=agegomp;k<=AGESUP;k++) {
10525: agemortsup=k;
10526: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10527: }
10528:
10529: for (k=agegomp;k<agemortsup;k++)
10530: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10531:
10532: for (k=agegomp;k<agemortsup;k++){
10533: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10534: sumlpop=sumlpop+lpop[k];
10535: }
10536:
10537: tpop[agegomp]=sumlpop;
10538: for (k=agegomp;k<(agemortsup-3);k++){
10539: /* tpop[k+1]=2;*/
10540: tpop[k+1]=tpop[k]-lpop[k];
10541: }
10542:
10543:
10544: printf("\nAge lx qx dx Lx Tx e(x)\n");
10545: for (k=agegomp;k<(agemortsup-2);k++)
10546: 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]);
10547:
10548:
10549: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10550: ageminpar=50;
10551: agemaxpar=100;
1.194 brouard 10552: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10553: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10554: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10555: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10556: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10557: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10558: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10559: }else{
10560: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10561: 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 10562: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10563: }
1.201 brouard 10564: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10565: stepm, weightopt,\
10566: model,imx,p,matcov,agemortsup);
10567:
10568: free_vector(lsurv,1,AGESUP);
10569: free_vector(lpop,1,AGESUP);
10570: free_vector(tpop,1,AGESUP);
1.220 brouard 10571: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10572: free_ivector(cens,1,n);
10573: free_vector(agecens,1,n);
10574: free_ivector(dcwave,1,n);
1.220 brouard 10575: #ifdef GSL
1.136 brouard 10576: #endif
1.186 brouard 10577: } /* Endof if mle==-3 mortality only */
1.205 brouard 10578: /* Standard */
10579: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10580: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10581: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10582: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10583: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10584: for (k=1; k<=npar;k++)
10585: printf(" %d %8.5f",k,p[k]);
10586: printf("\n");
1.205 brouard 10587: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10588: /* mlikeli uses func not funcone */
10589: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10590: }
10591: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10592: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10593: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10594: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10595: }
10596: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10597: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10598: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10599: for (k=1; k<=npar;k++)
10600: printf(" %d %8.5f",k,p[k]);
10601: printf("\n");
10602:
10603: /*--------- results files --------------*/
1.224 brouard 10604: 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 10605:
10606:
10607: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10608: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10609: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10610: for(i=1,jk=1; i <=nlstate; i++){
10611: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10612: if (k != i) {
10613: printf("%d%d ",i,k);
10614: fprintf(ficlog,"%d%d ",i,k);
10615: fprintf(ficres,"%1d%1d ",i,k);
10616: for(j=1; j <=ncovmodel; j++){
10617: printf("%12.7f ",p[jk]);
10618: fprintf(ficlog,"%12.7f ",p[jk]);
10619: fprintf(ficres,"%12.7f ",p[jk]);
10620: jk++;
10621: }
10622: printf("\n");
10623: fprintf(ficlog,"\n");
10624: fprintf(ficres,"\n");
10625: }
1.126 brouard 10626: }
10627: }
1.203 brouard 10628: if(mle != 0){
10629: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10630: ftolhess=ftol; /* Usually correct */
1.203 brouard 10631: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10632: 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");
10633: 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");
10634: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10635: for(k=1; k <=(nlstate+ndeath); k++){
10636: if (k != i) {
10637: printf("%d%d ",i,k);
10638: fprintf(ficlog,"%d%d ",i,k);
10639: for(j=1; j <=ncovmodel; j++){
10640: 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]));
10641: 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]));
10642: jk++;
10643: }
10644: printf("\n");
10645: fprintf(ficlog,"\n");
10646: }
10647: }
1.193 brouard 10648: }
1.203 brouard 10649: } /* end of hesscov and Wald tests */
1.225 brouard 10650:
1.203 brouard 10651: /* */
1.126 brouard 10652: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10653: printf("# Scales (for hessian or gradient estimation)\n");
10654: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10655: for(i=1,jk=1; i <=nlstate; i++){
10656: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10657: if (j!=i) {
10658: fprintf(ficres,"%1d%1d",i,j);
10659: printf("%1d%1d",i,j);
10660: fprintf(ficlog,"%1d%1d",i,j);
10661: for(k=1; k<=ncovmodel;k++){
10662: printf(" %.5e",delti[jk]);
10663: fprintf(ficlog," %.5e",delti[jk]);
10664: fprintf(ficres," %.5e",delti[jk]);
10665: jk++;
10666: }
10667: printf("\n");
10668: fprintf(ficlog,"\n");
10669: fprintf(ficres,"\n");
10670: }
1.126 brouard 10671: }
10672: }
10673:
10674: 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 10675: if(mle >= 1) /* To big for the screen */
1.126 brouard 10676: 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");
10677: 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");
10678: /* # 121 Var(a12)\n\ */
10679: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10680: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10681: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10682: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10683: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10684: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10685: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10686:
10687:
10688: /* Just to have a covariance matrix which will be more understandable
10689: even is we still don't want to manage dictionary of variables
10690: */
10691: for(itimes=1;itimes<=2;itimes++){
10692: jj=0;
10693: for(i=1; i <=nlstate; i++){
1.225 brouard 10694: for(j=1; j <=nlstate+ndeath; j++){
10695: if(j==i) continue;
10696: for(k=1; k<=ncovmodel;k++){
10697: jj++;
10698: ca[0]= k+'a'-1;ca[1]='\0';
10699: if(itimes==1){
10700: if(mle>=1)
10701: printf("#%1d%1d%d",i,j,k);
10702: fprintf(ficlog,"#%1d%1d%d",i,j,k);
10703: fprintf(ficres,"#%1d%1d%d",i,j,k);
10704: }else{
10705: if(mle>=1)
10706: printf("%1d%1d%d",i,j,k);
10707: fprintf(ficlog,"%1d%1d%d",i,j,k);
10708: fprintf(ficres,"%1d%1d%d",i,j,k);
10709: }
10710: ll=0;
10711: for(li=1;li <=nlstate; li++){
10712: for(lj=1;lj <=nlstate+ndeath; lj++){
10713: if(lj==li) continue;
10714: for(lk=1;lk<=ncovmodel;lk++){
10715: ll++;
10716: if(ll<=jj){
10717: cb[0]= lk +'a'-1;cb[1]='\0';
10718: if(ll<jj){
10719: if(itimes==1){
10720: if(mle>=1)
10721: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10722: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10723: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10724: }else{
10725: if(mle>=1)
10726: printf(" %.5e",matcov[jj][ll]);
10727: fprintf(ficlog," %.5e",matcov[jj][ll]);
10728: fprintf(ficres," %.5e",matcov[jj][ll]);
10729: }
10730: }else{
10731: if(itimes==1){
10732: if(mle>=1)
10733: printf(" Var(%s%1d%1d)",ca,i,j);
10734: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
10735: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
10736: }else{
10737: if(mle>=1)
10738: printf(" %.7e",matcov[jj][ll]);
10739: fprintf(ficlog," %.7e",matcov[jj][ll]);
10740: fprintf(ficres," %.7e",matcov[jj][ll]);
10741: }
10742: }
10743: }
10744: } /* end lk */
10745: } /* end lj */
10746: } /* end li */
10747: if(mle>=1)
10748: printf("\n");
10749: fprintf(ficlog,"\n");
10750: fprintf(ficres,"\n");
10751: numlinepar++;
10752: } /* end k*/
10753: } /*end j */
1.126 brouard 10754: } /* end i */
10755: } /* end itimes */
10756:
10757: fflush(ficlog);
10758: fflush(ficres);
1.225 brouard 10759: while(fgets(line, MAXLINE, ficpar)) {
10760: /* If line starts with a # it is a comment */
10761: if (line[0] == '#') {
10762: numlinepar++;
10763: fputs(line,stdout);
10764: fputs(line,ficparo);
10765: fputs(line,ficlog);
10766: continue;
10767: }else
10768: break;
10769: }
10770:
1.209 brouard 10771: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
10772: /* ungetc(c,ficpar); */
10773: /* fgets(line, MAXLINE, ficpar); */
10774: /* fputs(line,stdout); */
10775: /* fputs(line,ficparo); */
10776: /* } */
10777: /* ungetc(c,ficpar); */
1.126 brouard 10778:
10779: estepm=0;
1.209 brouard 10780: 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 10781:
10782: if (num_filled != 6) {
10783: 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);
10784: 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);
10785: goto end;
10786: }
10787: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
10788: }
10789: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
10790: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
10791:
1.209 brouard 10792: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 10793: if (estepm==0 || estepm < stepm) estepm=stepm;
10794: if (fage <= 2) {
10795: bage = ageminpar;
10796: fage = agemaxpar;
10797: }
10798:
10799: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 10800: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
10801: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 10802:
1.186 brouard 10803: /* Other stuffs, more or less useful */
1.126 brouard 10804: while((c=getc(ficpar))=='#' && c!= EOF){
10805: ungetc(c,ficpar);
10806: fgets(line, MAXLINE, ficpar);
1.141 brouard 10807: fputs(line,stdout);
1.126 brouard 10808: fputs(line,ficparo);
10809: }
10810: ungetc(c,ficpar);
10811:
10812: 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);
10813: 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);
10814: 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);
10815: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
10816: 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);
10817:
10818: while((c=getc(ficpar))=='#' && c!= EOF){
10819: ungetc(c,ficpar);
10820: fgets(line, MAXLINE, ficpar);
1.141 brouard 10821: fputs(line,stdout);
1.126 brouard 10822: fputs(line,ficparo);
10823: }
10824: ungetc(c,ficpar);
10825:
10826:
10827: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
10828: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
10829:
10830: fscanf(ficpar,"pop_based=%d\n",&popbased);
1.193 brouard 10831: fprintf(ficlog,"pop_based=%d\n",popbased);
1.126 brouard 10832: fprintf(ficparo,"pop_based=%d\n",popbased);
10833: fprintf(ficres,"pop_based=%d\n",popbased);
10834:
10835: while((c=getc(ficpar))=='#' && c!= EOF){
10836: ungetc(c,ficpar);
10837: fgets(line, MAXLINE, ficpar);
1.141 brouard 10838: fputs(line,stdout);
1.238 brouard 10839: fputs(line,ficres);
1.126 brouard 10840: fputs(line,ficparo);
10841: }
10842: ungetc(c,ficpar);
10843:
10844: 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);
10845: 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);
10846: 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);
10847: 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);
10848: 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);
10849: /* day and month of proj2 are not used but only year anproj2.*/
10850:
1.217 brouard 10851: while((c=getc(ficpar))=='#' && c!= EOF){
10852: ungetc(c,ficpar);
10853: fgets(line, MAXLINE, ficpar);
10854: fputs(line,stdout);
10855: fputs(line,ficparo);
1.238 brouard 10856: fputs(line,ficres);
1.217 brouard 10857: }
10858: ungetc(c,ficpar);
10859:
10860: 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 10861: 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);
10862: 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);
10863: 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 10864: /* day and month of proj2 are not used but only year anproj2.*/
1.126 brouard 10865:
1.230 brouard 10866: /* Results */
1.235 brouard 10867: nresult=0;
1.230 brouard 10868: while(fgets(line, MAXLINE, ficpar)) {
10869: /* If line starts with a # it is a comment */
10870: if (line[0] == '#') {
10871: numlinepar++;
10872: fputs(line,stdout);
10873: fputs(line,ficparo);
10874: fputs(line,ficlog);
1.238 brouard 10875: fputs(line,ficres);
1.230 brouard 10876: continue;
10877: }else
10878: break;
10879: }
1.240 brouard 10880: if (!feof(ficpar))
1.230 brouard 10881: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 10882: if (num_filled == 0){
1.230 brouard 10883: resultline[0]='\0';
1.240 brouard 10884: break;
10885: } else if (num_filled != 1){
1.230 brouard 10886: printf("ERROR %d: result line should be at minimum 'result=' %s\n",num_filled, line);
10887: }
1.235 brouard 10888: nresult++; /* Sum of resultlines */
10889: printf("Result %d: result=%s\n",nresult, resultline);
10890: if(nresult > MAXRESULTLINES){
10891: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
10892: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
10893: goto end;
10894: }
10895: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 10896: fprintf(ficparo,"result: %s\n",resultline);
10897: fprintf(ficres,"result: %s\n",resultline);
10898: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 10899: while(fgets(line, MAXLINE, ficpar)) {
10900: /* If line starts with a # it is a comment */
10901: if (line[0] == '#') {
10902: numlinepar++;
10903: fputs(line,stdout);
10904: fputs(line,ficparo);
1.238 brouard 10905: fputs(line,ficres);
1.230 brouard 10906: fputs(line,ficlog);
10907: continue;
10908: }else
10909: break;
10910: }
10911: if (feof(ficpar))
10912: break;
10913: else{ /* Processess output results for this combination of covariate values */
10914: }
1.240 brouard 10915: } /* end while */
1.230 brouard 10916:
10917:
1.126 brouard 10918:
1.230 brouard 10919: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 10920: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 10921:
10922: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 10923: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 10924: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 10925: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10926: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 10927: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 10928: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10929: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10930: }else{
1.218 brouard 10931: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 10932: }
10933: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 10934: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
10935: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 10936:
1.225 brouard 10937: /*------------ free_vector -------------*/
10938: /* chdir(path); */
1.220 brouard 10939:
1.215 brouard 10940: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
10941: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
10942: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
10943: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 10944: free_lvector(num,1,n);
10945: free_vector(agedc,1,n);
10946: /*free_matrix(covar,0,NCOVMAX,1,n);*/
10947: /*free_matrix(covar,1,NCOVMAX,1,n);*/
10948: fclose(ficparo);
10949: fclose(ficres);
1.220 brouard 10950:
10951:
1.186 brouard 10952: /* Other results (useful)*/
1.220 brouard 10953:
10954:
1.126 brouard 10955: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 10956: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
10957: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 10958: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 10959: fclose(ficrespl);
10960:
10961: /*------------- h Pij x at various ages ------------*/
1.180 brouard 10962: /*#include "hpijx.h"*/
10963: hPijx(p, bage, fage);
1.145 brouard 10964: fclose(ficrespij);
1.227 brouard 10965:
1.220 brouard 10966: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 10967: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 10968: k=1;
1.126 brouard 10969: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 10970:
1.219 brouard 10971: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 10972: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 10973: for(i=1;i<=AGESUP;i++)
1.219 brouard 10974: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 10975: for(k=1;k<=ncovcombmax;k++)
10976: probs[i][j][k]=0.;
1.219 brouard 10977: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
10978: if (mobilav!=0 ||mobilavproj !=0 ) {
10979: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 10980: for(i=1;i<=AGESUP;i++)
10981: for(j=1;j<=nlstate;j++)
10982: for(k=1;k<=ncovcombmax;k++)
10983: mobaverages[i][j][k]=0.;
1.219 brouard 10984: mobaverage=mobaverages;
10985: if (mobilav!=0) {
1.235 brouard 10986: printf("Movingaveraging observed prevalence\n");
1.227 brouard 10987: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
10988: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
10989: printf(" Error in movingaverage mobilav=%d\n",mobilav);
10990: }
1.219 brouard 10991: }
10992: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
10993: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
10994: else if (mobilavproj !=0) {
1.235 brouard 10995: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 10996: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
10997: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
10998: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
10999: }
1.219 brouard 11000: }
11001: }/* end if moving average */
1.227 brouard 11002:
1.126 brouard 11003: /*---------- Forecasting ------------------*/
11004: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11005: if(prevfcast==1){
11006: /* if(stepm ==1){*/
1.225 brouard 11007: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11008: }
1.217 brouard 11009: if(backcast==1){
1.219 brouard 11010: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11011: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11012: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11013:
11014: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11015:
11016: bprlim=matrix(1,nlstate,1,nlstate);
11017: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11018: fclose(ficresplb);
11019:
1.222 brouard 11020: hBijx(p, bage, fage, mobaverage);
11021: fclose(ficrespijb);
1.219 brouard 11022: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11023:
11024: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11025: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11026: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11027: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11028: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11029: }
1.217 brouard 11030:
1.186 brouard 11031:
11032: /* ------ Other prevalence ratios------------ */
1.126 brouard 11033:
1.215 brouard 11034: free_ivector(wav,1,imx);
11035: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11036: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11037: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11038:
11039:
1.127 brouard 11040: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11041:
1.201 brouard 11042: strcpy(filerese,"E_");
11043: strcat(filerese,fileresu);
1.126 brouard 11044: if((ficreseij=fopen(filerese,"w"))==NULL) {
11045: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11046: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11047: }
1.208 brouard 11048: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11049: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11050:
11051: pstamp(ficreseij);
1.219 brouard 11052:
1.235 brouard 11053: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11054: if (cptcovn < 1){i1=1;}
11055:
11056: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11057: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11058: if(TKresult[nres]!= k)
11059: continue;
1.219 brouard 11060: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11061: printf("\n#****** ");
1.225 brouard 11062: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11063: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11064: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11065: }
11066: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11067: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11068: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11069: }
11070: fprintf(ficreseij,"******\n");
1.235 brouard 11071: printf("******\n");
1.219 brouard 11072:
11073: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11074: oldm=oldms;savm=savms;
1.235 brouard 11075: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11076:
1.219 brouard 11077: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11078: }
11079: fclose(ficreseij);
1.208 brouard 11080: printf("done evsij\n");fflush(stdout);
11081: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11082:
1.227 brouard 11083: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11084:
11085:
1.201 brouard 11086: strcpy(filerest,"T_");
11087: strcat(filerest,fileresu);
1.127 brouard 11088: if((ficrest=fopen(filerest,"w"))==NULL) {
11089: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11090: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11091: }
1.208 brouard 11092: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11093: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11094:
1.126 brouard 11095:
1.201 brouard 11096: strcpy(fileresstde,"STDE_");
11097: strcat(fileresstde,fileresu);
1.126 brouard 11098: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11099: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11100: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11101: }
1.227 brouard 11102: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11103: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11104:
1.201 brouard 11105: strcpy(filerescve,"CVE_");
11106: strcat(filerescve,fileresu);
1.126 brouard 11107: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11108: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11109: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11110: }
1.227 brouard 11111: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11112: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11113:
1.201 brouard 11114: strcpy(fileresv,"V_");
11115: strcat(fileresv,fileresu);
1.126 brouard 11116: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11117: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11118: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11119: }
1.227 brouard 11120: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11121: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11122:
1.145 brouard 11123: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11124: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11125:
1.235 brouard 11126: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11127: if (cptcovn < 1){i1=1;}
11128:
11129: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11130: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11131: if(TKresult[nres]!= k)
11132: continue;
11133: printf("\n#****** Selected:");
11134: fprintf(ficrest,"\n#****** Selected:");
11135: fprintf(ficlog,"\n#****** Selected:");
1.227 brouard 11136: for(j=1;j<=cptcoveff;j++){
11137: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11138: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11139: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11140: }
1.235 brouard 11141: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11142: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11143: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11144: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11145: }
1.208 brouard 11146: fprintf(ficrest,"******\n");
1.227 brouard 11147: fprintf(ficlog,"******\n");
11148: printf("******\n");
1.208 brouard 11149:
11150: fprintf(ficresstdeij,"\n#****** ");
11151: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11152: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11153: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11154: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11155: }
1.235 brouard 11156: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11157: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11158: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11159: }
1.208 brouard 11160: fprintf(ficresstdeij,"******\n");
11161: fprintf(ficrescveij,"******\n");
11162:
11163: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11164: /* pstamp(ficresvij); */
1.225 brouard 11165: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11166: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11167: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11168: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11169: }
1.208 brouard 11170: fprintf(ficresvij,"******\n");
11171:
11172: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11173: oldm=oldms;savm=savms;
1.235 brouard 11174: printf(" cvevsij ");
11175: fprintf(ficlog, " cvevsij ");
11176: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11177: printf(" end cvevsij \n ");
11178: fprintf(ficlog, " end cvevsij \n ");
11179:
11180: /*
11181: */
11182: /* goto endfree; */
11183:
11184: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11185: pstamp(ficrest);
11186:
11187:
11188: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11189: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11190: cptcod= 0; /* To be deleted */
11191: printf("varevsij vpopbased=%d \n",vpopbased);
11192: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11193: 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 11194: 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 ");
11195: if(vpopbased==1)
11196: 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);
11197: else
11198: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11199: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11200: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11201: fprintf(ficrest,"\n");
11202: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11203: epj=vector(1,nlstate+1);
11204: printf("Computing age specific period (stable) prevalences in each health state \n");
11205: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11206: for(age=bage; age <=fage ;age++){
1.235 brouard 11207: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11208: if (vpopbased==1) {
11209: if(mobilav ==0){
11210: for(i=1; i<=nlstate;i++)
11211: prlim[i][i]=probs[(int)age][i][k];
11212: }else{ /* mobilav */
11213: for(i=1; i<=nlstate;i++)
11214: prlim[i][i]=mobaverage[(int)age][i][k];
11215: }
11216: }
1.219 brouard 11217:
1.227 brouard 11218: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11219: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11220: /* printf(" age %4.0f ",age); */
11221: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11222: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11223: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11224: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11225: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11226: }
11227: epj[nlstate+1] +=epj[j];
11228: }
11229: /* printf(" age %4.0f \n",age); */
1.219 brouard 11230:
1.227 brouard 11231: for(i=1, vepp=0.;i <=nlstate;i++)
11232: for(j=1;j <=nlstate;j++)
11233: vepp += vareij[i][j][(int)age];
11234: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11235: for(j=1;j <=nlstate;j++){
11236: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11237: }
11238: fprintf(ficrest,"\n");
11239: }
1.208 brouard 11240: } /* End vpopbased */
11241: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11242: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11243: free_vector(epj,1,nlstate+1);
1.235 brouard 11244: printf("done selection\n");fflush(stdout);
11245: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11246:
1.145 brouard 11247: /*}*/
1.235 brouard 11248: } /* End k selection */
1.227 brouard 11249:
11250: printf("done State-specific expectancies\n");fflush(stdout);
11251: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11252:
1.126 brouard 11253: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11254:
1.201 brouard 11255: strcpy(fileresvpl,"VPL_");
11256: strcat(fileresvpl,fileresu);
1.126 brouard 11257: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11258: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11259: exit(0);
11260: }
1.208 brouard 11261: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11262: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11263:
1.145 brouard 11264: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11265: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11266:
1.235 brouard 11267: i1=pow(2,cptcoveff);
11268: if (cptcovn < 1){i1=1;}
11269:
11270: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11271: for(k=1; k<=i1;k++){
11272: if(TKresult[nres]!= k)
11273: continue;
1.227 brouard 11274: fprintf(ficresvpl,"\n#****** ");
11275: printf("\n#****** ");
11276: fprintf(ficlog,"\n#****** ");
11277: for(j=1;j<=cptcoveff;j++) {
11278: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11279: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11280: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11281: }
1.235 brouard 11282: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11283: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11284: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11285: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11286: }
1.227 brouard 11287: fprintf(ficresvpl,"******\n");
11288: printf("******\n");
11289: fprintf(ficlog,"******\n");
11290:
11291: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11292: oldm=oldms;savm=savms;
1.235 brouard 11293: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11294: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11295: /*}*/
1.126 brouard 11296: }
1.227 brouard 11297:
1.126 brouard 11298: fclose(ficresvpl);
1.208 brouard 11299: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11300: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11301:
11302: free_vector(weight,1,n);
11303: free_imatrix(Tvard,1,NCOVMAX,1,2);
11304: free_imatrix(s,1,maxwav+1,1,n);
11305: free_matrix(anint,1,maxwav,1,n);
11306: free_matrix(mint,1,maxwav,1,n);
11307: free_ivector(cod,1,n);
11308: free_ivector(tab,1,NCOVMAX);
11309: fclose(ficresstdeij);
11310: fclose(ficrescveij);
11311: fclose(ficresvij);
11312: fclose(ficrest);
11313: fclose(ficpar);
11314:
11315:
1.126 brouard 11316: /*---------- End : free ----------------*/
1.219 brouard 11317: if (mobilav!=0 ||mobilavproj !=0)
11318: 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 11319: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11320: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11321: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11322: } /* mle==-3 arrives here for freeing */
1.227 brouard 11323: /* endfree:*/
11324: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11325: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11326: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11327: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11328: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11329: free_matrix(coqvar,1,maxwav,1,n);
11330: free_matrix(covar,0,NCOVMAX,1,n);
11331: free_matrix(matcov,1,npar,1,npar);
11332: free_matrix(hess,1,npar,1,npar);
11333: /*free_vector(delti,1,npar);*/
11334: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11335: free_matrix(agev,1,maxwav,1,imx);
11336: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11337:
11338: free_ivector(ncodemax,1,NCOVMAX);
11339: free_ivector(ncodemaxwundef,1,NCOVMAX);
11340: free_ivector(Dummy,-1,NCOVMAX);
11341: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11342: free_ivector(DummyV,1,NCOVMAX);
11343: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11344: free_ivector(Typevar,-1,NCOVMAX);
11345: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11346: free_ivector(TvarsQ,1,NCOVMAX);
11347: free_ivector(TvarsQind,1,NCOVMAX);
11348: free_ivector(TvarsD,1,NCOVMAX);
11349: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11350: free_ivector(TvarFD,1,NCOVMAX);
11351: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11352: free_ivector(TvarF,1,NCOVMAX);
11353: free_ivector(TvarFind,1,NCOVMAX);
11354: free_ivector(TvarV,1,NCOVMAX);
11355: free_ivector(TvarVind,1,NCOVMAX);
11356: free_ivector(TvarA,1,NCOVMAX);
11357: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11358: free_ivector(TvarFQ,1,NCOVMAX);
11359: free_ivector(TvarFQind,1,NCOVMAX);
11360: free_ivector(TvarVD,1,NCOVMAX);
11361: free_ivector(TvarVDind,1,NCOVMAX);
11362: free_ivector(TvarVQ,1,NCOVMAX);
11363: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11364: free_ivector(Tvarsel,1,NCOVMAX);
11365: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11366: free_ivector(Tposprod,1,NCOVMAX);
11367: free_ivector(Tprod,1,NCOVMAX);
11368: free_ivector(Tvaraff,1,NCOVMAX);
11369: free_ivector(invalidvarcomb,1,ncovcombmax);
11370: free_ivector(Tage,1,NCOVMAX);
11371: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11372: free_ivector(TmodelInvind,1,NCOVMAX);
11373: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11374:
11375: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11376: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11377: fflush(fichtm);
11378: fflush(ficgp);
11379:
1.227 brouard 11380:
1.126 brouard 11381: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11382: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11383: 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 11384: }else{
11385: printf("End of Imach\n");
11386: fprintf(ficlog,"End of Imach\n");
11387: }
11388: printf("See log file on %s\n",filelog);
11389: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11390: /*(void) gettimeofday(&end_time,&tzp);*/
11391: rend_time = time(NULL);
11392: end_time = *localtime(&rend_time);
11393: /* tml = *localtime(&end_time.tm_sec); */
11394: strcpy(strtend,asctime(&end_time));
1.126 brouard 11395: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11396: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11397: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11398:
1.157 brouard 11399: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11400: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11401: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11402: /* printf("Total time was %d uSec.\n", total_usecs);*/
11403: /* if(fileappend(fichtm,optionfilehtm)){ */
11404: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11405: fclose(fichtm);
11406: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11407: fclose(fichtmcov);
11408: fclose(ficgp);
11409: fclose(ficlog);
11410: /*------ End -----------*/
1.227 brouard 11411:
11412:
11413: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11414: #ifdef WIN32
1.227 brouard 11415: if (_chdir(pathcd) != 0)
11416: printf("Can't move to directory %s!\n",path);
11417: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11418: #else
1.227 brouard 11419: if(chdir(pathcd) != 0)
11420: printf("Can't move to directory %s!\n", path);
11421: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11422: #endif
1.126 brouard 11423: printf("Current directory %s!\n",pathcd);
11424: /*strcat(plotcmd,CHARSEPARATOR);*/
11425: sprintf(plotcmd,"gnuplot");
1.157 brouard 11426: #ifdef _WIN32
1.126 brouard 11427: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11428: #endif
11429: if(!stat(plotcmd,&info)){
1.158 brouard 11430: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11431: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11432: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11433: }else
11434: strcpy(pplotcmd,plotcmd);
1.157 brouard 11435: #ifdef __unix
1.126 brouard 11436: strcpy(plotcmd,GNUPLOTPROGRAM);
11437: if(!stat(plotcmd,&info)){
1.158 brouard 11438: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11439: }else
11440: strcpy(pplotcmd,plotcmd);
11441: #endif
11442: }else
11443: strcpy(pplotcmd,plotcmd);
11444:
11445: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11446: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11447:
1.126 brouard 11448: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11449: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11450: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11451: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11452: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11453: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11454: }
1.158 brouard 11455: printf(" Successful, please wait...");
1.126 brouard 11456: while (z[0] != 'q') {
11457: /* chdir(path); */
1.154 brouard 11458: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11459: scanf("%s",z);
11460: /* if (z[0] == 'c') system("./imach"); */
11461: if (z[0] == 'e') {
1.158 brouard 11462: #ifdef __APPLE__
1.152 brouard 11463: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11464: #elif __linux
11465: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11466: #else
1.152 brouard 11467: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11468: #endif
11469: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11470: system(pplotcmd);
1.126 brouard 11471: }
11472: else if (z[0] == 'g') system(plotcmd);
11473: else if (z[0] == 'q') exit(0);
11474: }
1.227 brouard 11475: end:
1.126 brouard 11476: while (z[0] != 'q') {
1.195 brouard 11477: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11478: scanf("%s",z);
11479: }
11480: }
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