1: 2015-08-03 Nicolas J Brouard <nbrouard@tugault.ined.fr>
2:
3: * (Module): Changing Variance of one-step probabilities into Standard deviation (thanks to Yao-Chi Shih)
4:
5: 2014-12-16 Nicolas J Brouard <nbrouard@tugault.local>
6:
7: * imach.c (Module): Merging 1.61 to 1.162
8:
9: 2014-09-02 Nicolas J Brouard <nbrouard@njbrouard.ined.fr>
10:
11: * prevlim.h (Module):
12:
13: 2014-01-26 Nicolas J Brouard <nbrouard@ledermann-2.local>
14:
15: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
16: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
17:
18: 2010-04-29 brouard <brouard@ined.fr>
19:
20: * imach.c (Module): Checking covariates for more complex models
21: than V1+V2. A lot of change to be done. Unstable.
22:
23: 2010-04-26 brouard <brouard@ined.fr>
24:
25: * imach.c (Module): merging some libgsl code. Fixing computation
26: of likelione (using inter/intrapolation if mle = 0) in order to
27: get same likelihood as if mle=1.
28: Some cleaning of code and comments added.
29:
30: 2009-10-29 brouard <brouard@ined.fr>
31:
32: * imach.c (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
33:
34: 2006-06-30 Brouard Nicolas <brouard@localhost>
35:
36: * imach.c (Module): Clarifications on computing e.j
37:
38: 2006-04-28 Brouard Nicolas <brouard@localhost>
39:
40: * imach.c (Module): Yes the sum of survivors was wrong since
41: imach-114 because nhstepm was no more computed in the age
42: loop. Now we define nhstepma in the age loop.
43: (Module): In order to speed up (in case of numerous covariates) we
44: compute health expectancies (without variances) in a first step
45: and then all the health expectancies with variances or standard
46: deviation (needs data from the Hessian matrices) which slows the
47: computation.
48: In the future we should be able to stop the program is only health
49: expectancies and graph are needed without standard deviations.
50:
51: 2006-04-04 LIEVRE Agnes <lievre@PC_00697.ined.fr>
52:
53: * imach.iss (Module):
54:
55: * imach.c (Module): bugs in evsi + population file added in htm output file
56:
57: 2006-03-22 LIEVRE Agnes <lievre@PC_00697.ined.fr>
58:
59: * imach.c (Module):
60:
61: 2006-03-20 Brouard Nicolas <brouard@localhost>
62:
63: * imach.c (Module): <title> changed, corresponds to .htm file
64: name. <head> headers where missing.
65:
66: * Makefile (Module): 0.98g
67:
68: * imach.c (Module): Weights can have a decimal point as for
69: English (a comma might work with a correct LC_NUMERIC environment,
70: otherwise the weight is truncated).
71: Modification of warning when the covariates values are not 0 or
72: 1.
73: Version 0.98g
74:
75: 2006-03-16 LIEVRE Agnes <lievre@PC_00697.ined.fr>
76:
77: * imach.c (Module): Comments concerning covariates added
78:
79: * noreg-setup.iss (Module): Version 0.98f
80:
81: * Makefile (Module): Version 0.98f make _windows=1 imachcyg-setup
82:
83: * imach.c (Module): refinements in the computation of lli if
84: status=-2 in order to have more reliable computation if stepm is
85: not 1 month. Version 0.98f
86:
87: 2006-03-15 Brouard Nicolas <brouard@localhost>
88:
89: * imach.c (Module): Bug if status = -2, the loglikelihood was
90: computed as likelihood omitting the logarithm. Version O.98e
91:
92: 2006-03-14 Brouard Nicolas <brouard@localhost>
93:
94: * imach.c (Module): varevsij Comments added explaining the second
95: table of variances if popbased=1 .
96: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
97: (Module): Function pstamp added
98: (Module): Version 0.98d
99:
100: 2006-03-06 Brouard Nicolas <brouard@localhost>
101:
102: * imach.c (Module): Variance-covariance wrong links and
103: varian-covariance of ej. is needed (Saito).
104:
105: 2006-02-27 Brouard Nicolas <brouard@localhost>
106:
107: * imach.c (Module): One freematrix added in mlikeli!
108:
109: 2006-02-26 Brouard Nicolas <brouard@localhost>
110:
111: * imach.c (Module): Some improvements in processing parameter
112: filename with strsep.
113:
114: 2006-02-24 Brouard Nicolas <brouard@localhost>
115:
116: * imach.c (Module): Memory leaks checks with valgrind and:
117: datafile was not closed, some imatrix were not freed and on matrix
118: allocation too.
119:
120: 2006-02-09 LIEVRE Agnes <lievre@PC_00697.ined.fr>
121:
122: * strsep.c (Module): Inclusion because not in mingw32!!!
123:
124: * setup.iss.in (Module): IMACHSETUPVERSION added
125:
126: * noreg-setup.iss (Module): added to cvs tree
127:
128: * Makefile (Module): IMACHSETUPVERSION added
129:
130: 2006-01-30 BROUARD Nicolas <brouard@localhost>
131:
132: * Makefile (Module, Module): Latest Makefile
133:
134: * imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe
135:
136: 2006-01-27 BROUARD Nicolas <brouard@localhost>
137:
138: * noreg-setup.iss (Module): ADDED
139:
140: 2006-01-24 BROUARD Nicolas <brouard@localhost>
141:
142: * imach.c (Module): Comments (lines starting with a #) are allowed in data.
143:
144: 2005-10-25 Nicolas Brouard <brouard@dhcp171.recherche.ined.fr>
145:
146: * Makefile (Module): Added Imach98.pmsp and howto crete a dmg
147:
148: * createdmg.sh (Module): Creates a dmg disk for MaC OS/X
149:
150: 2005-09-30 <lievre@PC_00595>
151:
152: * imach.c (Module): sump fixed, loop imx fixed, and simplifications.
153: (Module): If the status is missing at the last wave but we know
154: that the person is alive, then we can code his/her status as -2
155: (instead of missing=-1 in earlier versions) and his/her
156: contributions to the likelihood is 1 - Prob of dying from last
157: health status (= 1-p13= p11+p12 in the easiest case of somebody in
158: the healthy state at last known wave). Version is 0.98
159:
160: 2004-05-20 Brouard Nicolas <brouard@localhost>
161:
162: * imach.c (Repository):
163: Agnes added a direct estimation of mortality (without the need of
164: computing period prevalence and differential mortality). Thus here
165: is version 0.97a which has been distributed to some people at
166: REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of
167: mortality using covariates is not done today. Estimating direct
168: mortality is a very different process because it doesn't need
169: interpolation because it is easy to get the lx from the force of
170: the mortality mux in the simplest case as for a Gompertz (log mux
171: = a + b*x . But we have been able to incorporate the new code
172: within former imach program (0.96d) without deteriorating too much
173: the understanding of the program.
174:
175: 2003-06-25 <brouard@BROUARD>
176:
177: * imach.c (Module): On windows (cygwin) function asctime_r doesn't
178: exist so I changed back to asctime which exists.
179: (Module): Version 0.96b
180:
181: 2003-06-24 <brouard@BROUARD>
182:
183: * imach.c (Module): Some bugs corrected for windows. Also, when
184: mle=-1 a template is output in file "or"mypar.txt with the design
185: of the covariance matrix to be input.
186:
187: 2003-06-25 Brouard Nicolas <brouard@homebrou>
188:
189: * imach.c (Repository): Duplicated warning errors corrected.
190: (Repository): Elapsed time after each iteration is now output. It
191: helps to forecast when convergence will be reached. Elapsed time
192: is stamped in powell. We created a new html file for the graphs
193: concerning matrix of covariance. It has extension -cov.htm.
194:
195: 2003-06-23 Brouard Nicolas <brouard@homebrou>
196:
197: * 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.
198:
199: 2003-06-18 Brouard Nicolas <brouard@homebrou>
200:
201: * imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96
202:
203: 2003-06-17 Brouard Nicolas <brouard@homebrou>
204:
205:
206: * imach.c (Repository): Check when date of death was earlier that
207: current date of interview. It may happen when the death was just
208: prior to the death. In this case, dh was negative and likelihood
209: was wrong (infinity). We still send an "Error" but patch by
210: assuming that the date of death was just one stepm after the
211: interview.
212: (Repository): Because some people have very long ID (first column)
213: we changed int to long in num[] and we added a new lvector for
214: memory allocation. But we also truncated to 8 characters (left
215: truncation)
216:
217: (Repository): No more line truncation errors.
218:
219: 2003-06-14 Brouard Nicolas <brouard@homebrou>
220:
221: * imach.c (Repository): Add of a routine likelione (likelihood
222: only once) which prints on a text file
223: (ilk) the contributions to the likelihood for each
224: individual/wave.
225:
226: 2003-06-13 Brouard Nicolas <brouard@homebrou>
227: * imach.c (Repository): Replace "freqsummary" at a correct
228: place. It differs from routine "prevalence" which may be called
229: many times. Probs is memory consuming and must be used with
230: parcimony.
231:
232: 2003-06-17 <brouard@BROUARD>
233:
234: * timeval.h (Module): Added included file to make use of
235: gettimeofday working on win32 with cygwin.
236:
237: 2003-05-16 Brouard Nicolas <brouard@brouard>
238: * imach.c (Module): Insert of a warning if the delay between two
239: waves is negative.
240:
241: * imach.c (Module):
242:
243: 2003-05-03 <brouard@BROUARD>
244:
245: * imach.c: Some cleaning
246:
247: 2003-03-28 <brouard@BROUARD>
248: * imach.c (Module): In version up to 0.92 likelihood was computed
249: as if date of death was unknown. Death was treated as any other
250: health state: the date of the interview describes the actual state
251: and not the date of a change in health state. The former idea was
252: to consider that at each interview the state was recorded
253: (healthy, disable or death) and IMaCh was corrected; but when we
254: introduced the exact date of death then we should have modified
255: the contribution of an exact death to the likelihood. This new
256: contribution is smaller and very dependent of the step unit
257: stepm. It is no more the probability to die between last interview
258: and month of death but the probability to survive from last
259: interview up to one month before death multiplied by the
260: probability to die within a month. Thanks to Chris
261: Jackson for correcting this bug. Former versions increased
262: mortality artificially. The bad side is that we add another loop
263:
264:
265: which slows down the processing. The difference can be up to 10%
266:
267:
268: lower mortality.
269:
270:
271:
272:
273:
274: 2002-05-30 <brouard@BROUARD>
275:
276:
277:
278:
279:
280: *
281:
282:
283:
284:
285:
286: * imach.c (Module): Add correlation matrix of one-step
287:
288:
289: probabilities and covariance matrix
290:
291:
292:
293:
294:
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