Annotation of imach/src/ChangeLog, revision 1.37
1.37 ! brouard 1: 2015-09-15 Nicolas J Brouard <nbrouard@tugault.ined.fr>
! 2:
! 3: * (Module): 0.98r0 Some new graphs, some bugs fixed: model=1+age+V2. etc
! 4:
1.36 brouard 5: 2015-08-18 Nicolas J Brouard <nbrouard@tugault.local>
6:
7: * (Module): Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
8:
1.35 brouard 9: 2015-08-03 Nicolas J Brouard <nbrouard@tugault.ined.fr>
10:
11: * (Module): Changing Variance of one-step probabilities into Standard deviation (thanks to Yao-Chi Shih)
12:
13: 2014-12-16 Nicolas J Brouard <nbrouard@tugault.local>
14:
15: * imach.c (Module): Merging 1.61 to 1.162
16:
17: 2014-09-02 Nicolas J Brouard <nbrouard@njbrouard.ined.fr>
18:
19: * prevlim.h (Module):
20:
1.34 brouard 21: 2014-01-26 Nicolas J Brouard <nbrouard@ledermann-2.local>
22:
23: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
24: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
25:
1.33 brouard 26: 2010-04-29 brouard <brouard@ined.fr>
27:
28: * imach.c (Module): Checking covariates for more complex models
29: than V1+V2. A lot of change to be done. Unstable.
30:
31: 2010-04-26 brouard <brouard@ined.fr>
32:
33: * imach.c (Module): merging some libgsl code. Fixing computation
34: of likelione (using inter/intrapolation if mle = 0) in order to
35: get same likelihood as if mle=1.
36: Some cleaning of code and comments added.
37:
1.32 brouard 38: 2009-10-29 brouard <brouard@ined.fr>
39:
40: * imach.c (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
41:
1.31 brouard 42: 2006-06-30 Brouard Nicolas <brouard@localhost>
43:
44: * imach.c (Module): Clarifications on computing e.j
45:
1.30 brouard 46: 2006-04-28 Brouard Nicolas <brouard@localhost>
47:
48: * imach.c (Module): Yes the sum of survivors was wrong since
49: imach-114 because nhstepm was no more computed in the age
50: loop. Now we define nhstepma in the age loop.
1.31 brouard 51: (Module): In order to speed up (in case of numerous covariates) we
52: compute health expectancies (without variances) in a first step
53: and then all the health expectancies with variances or standard
54: deviation (needs data from the Hessian matrices) which slows the
55: computation.
56: In the future we should be able to stop the program is only health
57: expectancies and graph are needed without standard deviations.
1.30 brouard 58:
1.29 lievre 59: 2006-04-04 LIEVRE Agnes <lievre@PC_00697.ined.fr>
60:
61: * imach.iss (Module):
62:
63: * imach.c (Module): bugs in evsi + population file added in htm output file
64:
65: 2006-03-22 LIEVRE Agnes <lievre@PC_00697.ined.fr>
66:
67: * imach.c (Module):
68:
1.27 brouard 69: 2006-03-20 Brouard Nicolas <brouard@localhost>
70:
1.28 brouard 71: * imach.c (Module): <title> changed, corresponds to .htm file
72: name. <head> headers where missing.
73:
1.27 brouard 74: * Makefile (Module): 0.98g
75:
76: * imach.c (Module): Weights can have a decimal point as for
77: English (a comma might work with a correct LC_NUMERIC environment,
78: otherwise the weight is truncated).
79: Modification of warning when the covariates values are not 0 or
80: 1.
81: Version 0.98g
82:
1.24 lievre 83: 2006-03-16 LIEVRE Agnes <lievre@PC_00697.ined.fr>
84:
1.26 lievre 85: * imach.c (Module): Comments concerning covariates added
86:
1.25 lievre 87: * noreg-setup.iss (Module): Version 0.98f
88:
1.24 lievre 89: * Makefile (Module): Version 0.98f make _windows=1 imachcyg-setup
90:
91: * imach.c (Module): refinements in the computation of lli if
92: status=-2 in order to have more reliable computation if stepm is
93: not 1 month. Version 0.98f
94:
1.22 brouard 95: 2006-03-15 Brouard Nicolas <brouard@localhost>
96:
97: * imach.c (Module): Bug if status = -2, the loglikelihood was
98: computed as likelihood omitting the logarithm. Version O.98e
99:
1.21 brouard 100: 2006-03-14 Brouard Nicolas <brouard@localhost>
101:
102: * imach.c (Module): varevsij Comments added explaining the second
103: table of variances if popbased=1 .
104: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
105: (Module): Function pstamp added
106: (Module): Version 0.98d
107:
1.20 brouard 108: 2006-03-06 Brouard Nicolas <brouard@localhost>
109:
110: * imach.c (Module): Variance-covariance wrong links and
111: varian-covariance of ej. is needed (Saito).
112:
1.19 brouard 113: 2006-02-27 Brouard Nicolas <brouard@localhost>
114:
115: * imach.c (Module): One freematrix added in mlikeli!
116:
117: 2006-02-26 Brouard Nicolas <brouard@localhost>
118:
119: * imach.c (Module): Some improvements in processing parameter
120: filename with strsep.
121:
122: 2006-02-24 Brouard Nicolas <brouard@localhost>
123:
124: * imach.c (Module): Memory leaks checks with valgrind and:
125: datafile was not closed, some imatrix were not freed and on matrix
126: allocation too.
127:
1.16 lievre 128: 2006-02-09 LIEVRE Agnes <lievre@PC_00697.ined.fr>
129:
1.18 lievre 130: * strsep.c (Module): Inclusion because not in mingw32!!!
131:
1.17 lievre 132: * setup.iss.in (Module): IMACHSETUPVERSION added
133:
134: * noreg-setup.iss (Module): added to cvs tree
135:
1.16 lievre 136: * Makefile (Module): IMACHSETUPVERSION added
137:
1.15 brouard 138: 2006-01-30 BROUARD Nicolas <brouard@localhost>
139:
1.23 brouard 140: * Makefile (Module, Module): Latest Makefile
141:
1.15 brouard 142: * imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe
143:
144: 2006-01-27 BROUARD Nicolas <brouard@localhost>
145:
146: * noreg-setup.iss (Module): ADDED
147:
148: 2006-01-24 BROUARD Nicolas <brouard@localhost>
149:
150: * imach.c (Module): Comments (lines starting with a #) are allowed in data.
151:
1.14 brouard 152: 2005-10-25 Nicolas Brouard <brouard@dhcp171.recherche.ined.fr>
153:
154: * Makefile (Module): Added Imach98.pmsp and howto crete a dmg
155:
156: * createdmg.sh (Module): Creates a dmg disk for MaC OS/X
157:
1.12 lievre 158: 2005-09-30 <lievre@PC_00595>
159:
160: * imach.c (Module): sump fixed, loop imx fixed, and simplifications.
1.13 lievre 161: (Module): If the status is missing at the last wave but we know
162: that the person is alive, then we can code his/her status as -2
163: (instead of missing=-1 in earlier versions) and his/her
164: contributions to the likelihood is 1 - Prob of dying from last
165: health status (= 1-p13= p11+p12 in the easiest case of somebody in
166: the healthy state at last known wave). Version is 0.98
1.12 lievre 167:
1.11 brouard 168: 2004-05-20 Brouard Nicolas <brouard@localhost>
169:
170: * imach.c (Repository):
171: Agnes added a direct estimation of mortality (without the need of
172: computing period prevalence and differential mortality). Thus here
173: is version 0.97a which has been distributed to some people at
174: REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of
175: mortality using covariates is not done today. Estimating direct
176: mortality is a very different process because it doesn't need
177: interpolation because it is easy to get the lx from the force of
178: the mortality mux in the simplest case as for a Gompertz (log mux
179: = a + b*x . But we have been able to incorporate the new code
180: within former imach program (0.96d) without deteriorating too much
181: the understanding of the program.
182:
1.10 brouard 183: 2003-06-25 <brouard@BROUARD>
184:
185: * imach.c (Module): On windows (cygwin) function asctime_r doesn't
186: exist so I changed back to asctime which exists.
187: (Module): Version 0.96b
188:
189: 2003-06-24 <brouard@BROUARD>
190:
191: * imach.c (Module): Some bugs corrected for windows. Also, when
192: mle=-1 a template is output in file "or"mypar.txt with the design
193: of the covariance matrix to be input.
194:
1.9 brouard 195: 2003-06-25 Brouard Nicolas <brouard@homebrou>
196:
197: * imach.c (Repository): Duplicated warning errors corrected.
198: (Repository): Elapsed time after each iteration is now output. It
199: helps to forecast when convergence will be reached. Elapsed time
200: is stamped in powell. We created a new html file for the graphs
201: concerning matrix of covariance. It has extension -cov.htm.
202:
1.8 brouard 203: 2003-06-23 Brouard Nicolas <brouard@homebrou>
204:
205: * 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.
206:
207: 2003-06-18 Brouard Nicolas <brouard@homebrou>
208:
209: * imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96
210:
1.7 brouard 211: 2003-06-17 Brouard Nicolas <brouard@homebrou>
212:
213:
214: * imach.c (Repository): Check when date of death was earlier that
215: current date of interview. It may happen when the death was just
216: prior to the death. In this case, dh was negative and likelihood
217: was wrong (infinity). We still send an "Error" but patch by
218: assuming that the date of death was just one stepm after the
219: interview.
220: (Repository): Because some people have very long ID (first column)
221: we changed int to long in num[] and we added a new lvector for
222: memory allocation. But we also truncated to 8 characters (left
223: truncation)
224:
225: (Repository): No more line truncation errors.
226:
227: 2003-06-14 Brouard Nicolas <brouard@homebrou>
228:
229: * imach.c (Repository): Add of a routine likelione (likelihood
230: only once) which prints on a text file
231: (ilk) the contributions to the likelihood for each
232: individual/wave.
233:
234: 2003-06-13 Brouard Nicolas <brouard@homebrou>
235: * imach.c (Repository): Replace "freqsummary" at a correct
236: place. It differs from routine "prevalence" which may be called
237: many times. Probs is memory consuming and must be used with
238: parcimony.
239:
240: 2003-06-17 <brouard@BROUARD>
241:
242: * timeval.h (Module): Added included file to make use of
243: gettimeofday working on win32 with cygwin.
244:
245: 2003-05-16 Brouard Nicolas <brouard@brouard>
246: * imach.c (Module): Insert of a warning if the delay between two
247: waves is negative.
248:
249: * imach.c (Module):
250:
251: 2003-05-03 <brouard@BROUARD>
252:
253: * imach.c: Some cleaning
254:
255: 2003-03-28 <brouard@BROUARD>
256: * imach.c (Module): In version up to 0.92 likelihood was computed
257: as if date of death was unknown. Death was treated as any other
258: health state: the date of the interview describes the actual state
259: and not the date of a change in health state. The former idea was
260: to consider that at each interview the state was recorded
261: (healthy, disable or death) and IMaCh was corrected; but when we
262: introduced the exact date of death then we should have modified
263: the contribution of an exact death to the likelihood. This new
264: contribution is smaller and very dependent of the step unit
265: stepm. It is no more the probability to die between last interview
266: and month of death but the probability to survive from last
267: interview up to one month before death multiplied by the
268: probability to die within a month. Thanks to Chris
269: Jackson for correcting this bug. Former versions increased
270: mortality artificially. The bad side is that we add another loop
271:
272:
273: which slows down the processing. The difference can be up to 10%
274:
275:
276: lower mortality.
277:
278:
279:
280:
281:
282: 2002-05-30 <brouard@BROUARD>
283:
284:
285:
286:
287:
288: *
289:
290:
291:
292:
293:
294: * imach.c (Module): Add correlation matrix of one-step
295:
296:
297: probabilities and covariance matrix
298:
299:
300:
301:
302:
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/icons/small/back.gif){kind=icon}
Return to ChangeLog CVS log ![[TXT]](/icons/small/text.gif){kind=icon}
![[DIR]](/icons/small/dir.gif){kind=icon}
Up to [Local Repository] / imach / src