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