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