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