1: 2006-06-30 Brouard Nicolas <brouard@localhost>
2:
3: * imach.c (Module): Clarifications on computing e.j
4:
5: 2006-04-28 Brouard Nicolas <brouard@localhost>
6:
7: * imach.c (Module): Yes the sum of survivors was wrong since
8: imach-114 because nhstepm was no more computed in the age
9: loop. Now we define nhstepma in the age loop.
10: (Module): In order to speed up (in case of numerous covariates) we
11: compute health expectancies (without variances) in a first step
12: and then all the health expectancies with variances or standard
13: deviation (needs data from the Hessian matrices) which slows the
14: computation.
15: In the future we should be able to stop the program is only health
16: expectancies and graph are needed without standard deviations.
17:
18: 2006-04-04 LIEVRE Agnes <lievre@PC_00697.ined.fr>
19:
20: * imach.iss (Module):
21:
22: * imach.c (Module): bugs in evsi + population file added in htm output file
23:
24: 2006-03-22 LIEVRE Agnes <lievre@PC_00697.ined.fr>
25:
26: * imach.c (Module):
27:
28: 2006-03-20 Brouard Nicolas <brouard@localhost>
29:
30: * imach.c (Module): <title> changed, corresponds to .htm file
31: name. <head> headers where missing.
32:
33: * Makefile (Module): 0.98g
34:
35: * imach.c (Module): Weights can have a decimal point as for
36: English (a comma might work with a correct LC_NUMERIC environment,
37: otherwise the weight is truncated).
38: Modification of warning when the covariates values are not 0 or
39: 1.
40: Version 0.98g
41:
42: 2006-03-16 LIEVRE Agnes <lievre@PC_00697.ined.fr>
43:
44: * imach.c (Module): Comments concerning covariates added
45:
46: * noreg-setup.iss (Module): Version 0.98f
47:
48: * Makefile (Module): Version 0.98f make _windows=1 imachcyg-setup
49:
50: * imach.c (Module): refinements in the computation of lli if
51: status=-2 in order to have more reliable computation if stepm is
52: not 1 month. Version 0.98f
53:
54: 2006-03-15 Brouard Nicolas <brouard@localhost>
55:
56: * imach.c (Module): Bug if status = -2, the loglikelihood was
57: computed as likelihood omitting the logarithm. Version O.98e
58:
59: 2006-03-14 Brouard Nicolas <brouard@localhost>
60:
61: * imach.c (Module): varevsij Comments added explaining the second
62: table of variances if popbased=1 .
63: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
64: (Module): Function pstamp added
65: (Module): Version 0.98d
66:
67: 2006-03-06 Brouard Nicolas <brouard@localhost>
68:
69: * imach.c (Module): Variance-covariance wrong links and
70: varian-covariance of ej. is needed (Saito).
71:
72: 2006-02-27 Brouard Nicolas <brouard@localhost>
73:
74: * imach.c (Module): One freematrix added in mlikeli!
75:
76: 2006-02-26 Brouard Nicolas <brouard@localhost>
77:
78: * imach.c (Module): Some improvements in processing parameter
79: filename with strsep.
80:
81: 2006-02-24 Brouard Nicolas <brouard@localhost>
82:
83: * imach.c (Module): Memory leaks checks with valgrind and:
84: datafile was not closed, some imatrix were not freed and on matrix
85: allocation too.
86:
87: 2006-02-09 LIEVRE Agnes <lievre@PC_00697.ined.fr>
88:
89: * strsep.c (Module): Inclusion because not in mingw32!!!
90:
91: * setup.iss.in (Module): IMACHSETUPVERSION added
92:
93: * noreg-setup.iss (Module): added to cvs tree
94:
95: * Makefile (Module): IMACHSETUPVERSION added
96:
97: 2006-01-30 BROUARD Nicolas <brouard@localhost>
98:
99: * Makefile (Module, Module): Latest Makefile
100:
101: * imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe
102:
103: 2006-01-27 BROUARD Nicolas <brouard@localhost>
104:
105: * noreg-setup.iss (Module): ADDED
106:
107: 2006-01-24 BROUARD Nicolas <brouard@localhost>
108:
109: * imach.c (Module): Comments (lines starting with a #) are allowed in data.
110:
111: 2005-10-25 Nicolas Brouard <brouard@dhcp171.recherche.ined.fr>
112:
113: * Makefile (Module): Added Imach98.pmsp and howto crete a dmg
114:
115: * createdmg.sh (Module): Creates a dmg disk for MaC OS/X
116:
117: 2005-09-30 <lievre@PC_00595>
118:
119: * imach.c (Module): sump fixed, loop imx fixed, and simplifications.
120: (Module): If the status is missing at the last wave but we know
121: that the person is alive, then we can code his/her status as -2
122: (instead of missing=-1 in earlier versions) and his/her
123: contributions to the likelihood is 1 - Prob of dying from last
124: health status (= 1-p13= p11+p12 in the easiest case of somebody in
125: the healthy state at last known wave). Version is 0.98
126:
127: 2004-05-20 Brouard Nicolas <brouard@localhost>
128:
129: * imach.c (Repository):
130: Agnes added a direct estimation of mortality (without the need of
131: computing period prevalence and differential mortality). Thus here
132: is version 0.97a which has been distributed to some people at
133: REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of
134: mortality using covariates is not done today. Estimating direct
135: mortality is a very different process because it doesn't need
136: interpolation because it is easy to get the lx from the force of
137: the mortality mux in the simplest case as for a Gompertz (log mux
138: = a + b*x . But we have been able to incorporate the new code
139: within former imach program (0.96d) without deteriorating too much
140: the understanding of the program.
141:
142: 2003-06-25 <brouard@BROUARD>
143:
144: * imach.c (Module): On windows (cygwin) function asctime_r doesn't
145: exist so I changed back to asctime which exists.
146: (Module): Version 0.96b
147:
148: 2003-06-24 <brouard@BROUARD>
149:
150: * imach.c (Module): Some bugs corrected for windows. Also, when
151: mle=-1 a template is output in file "or"mypar.txt with the design
152: of the covariance matrix to be input.
153:
154: 2003-06-25 Brouard Nicolas <brouard@homebrou>
155:
156: * imach.c (Repository): Duplicated warning errors corrected.
157: (Repository): Elapsed time after each iteration is now output. It
158: helps to forecast when convergence will be reached. Elapsed time
159: is stamped in powell. We created a new html file for the graphs
160: concerning matrix of covariance. It has extension -cov.htm.
161:
162: 2003-06-23 Brouard Nicolas <brouard@homebrou>
163:
164: * 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.
165:
166: 2003-06-18 Brouard Nicolas <brouard@homebrou>
167:
168: * imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96
169:
170: 2003-06-17 Brouard Nicolas <brouard@homebrou>
171:
172:
173: * imach.c (Repository): Check when date of death was earlier that
174: current date of interview. It may happen when the death was just
175: prior to the death. In this case, dh was negative and likelihood
176: was wrong (infinity). We still send an "Error" but patch by
177: assuming that the date of death was just one stepm after the
178: interview.
179: (Repository): Because some people have very long ID (first column)
180: we changed int to long in num[] and we added a new lvector for
181: memory allocation. But we also truncated to 8 characters (left
182: truncation)
183:
184: (Repository): No more line truncation errors.
185:
186: 2003-06-14 Brouard Nicolas <brouard@homebrou>
187:
188: * imach.c (Repository): Add of a routine likelione (likelihood
189: only once) which prints on a text file
190: (ilk) the contributions to the likelihood for each
191: individual/wave.
192:
193: 2003-06-13 Brouard Nicolas <brouard@homebrou>
194: * imach.c (Repository): Replace "freqsummary" at a correct
195: place. It differs from routine "prevalence" which may be called
196: many times. Probs is memory consuming and must be used with
197: parcimony.
198:
199: 2003-06-17 <brouard@BROUARD>
200:
201: * timeval.h (Module): Added included file to make use of
202: gettimeofday working on win32 with cygwin.
203:
204: 2003-05-16 Brouard Nicolas <brouard@brouard>
205: * imach.c (Module): Insert of a warning if the delay between two
206: waves is negative.
207:
208: * imach.c (Module):
209:
210: 2003-05-03 <brouard@BROUARD>
211:
212: * imach.c: Some cleaning
213:
214: 2003-03-28 <brouard@BROUARD>
215: * imach.c (Module): In version up to 0.92 likelihood was computed
216: as if date of death was unknown. Death was treated as any other
217: health state: the date of the interview describes the actual state
218: and not the date of a change in health state. The former idea was
219: to consider that at each interview the state was recorded
220: (healthy, disable or death) and IMaCh was corrected; but when we
221: introduced the exact date of death then we should have modified
222: the contribution of an exact death to the likelihood. This new
223: contribution is smaller and very dependent of the step unit
224: stepm. It is no more the probability to die between last interview
225: and month of death but the probability to survive from last
226: interview up to one month before death multiplied by the
227: probability to die within a month. Thanks to Chris
228: Jackson for correcting this bug. Former versions increased
229: mortality artificially. The bad side is that we add another loop
230:
231:
232: which slows down the processing. The difference can be up to 10%
233:
234:
235: lower mortality.
236:
237:
238:
239:
240:
241: 2002-05-30 <brouard@BROUARD>
242:
243:
244:
245:
246:
247: *
248:
249:
250:
251:
252:
253: * imach.c (Module): Add correlation matrix of one-step
254:
255:
256: probabilities and covariance matrix
257:
258:
259:
260:
261:
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