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