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