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