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