1: 2015-08-18 Nicolas J Brouard <nbrouard@tugault.local>
2:
3: * (Module): Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
4:
5: 2015-08-03 Nicolas J Brouard <nbrouard@tugault.ined.fr>
6:
7: * (Module): Changing Variance of one-step probabilities into Standard deviation (thanks to Yao-Chi Shih)
8:
9: 2014-12-16 Nicolas J Brouard <nbrouard@tugault.local>
10:
11: * imach.c (Module): Merging 1.61 to 1.162
12:
13: 2014-09-02 Nicolas J Brouard <nbrouard@njbrouard.ined.fr>
14:
15: * prevlim.h (Module):
16:
17: 2014-01-26 Nicolas J Brouard <nbrouard@ledermann-2.local>
18:
19: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
20: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
21:
22: 2010-04-29 brouard <brouard@ined.fr>
23:
24: * imach.c (Module): Checking covariates for more complex models
25: than V1+V2. A lot of change to be done. Unstable.
26:
27: 2010-04-26 brouard <brouard@ined.fr>
28:
29: * imach.c (Module): merging some libgsl code. Fixing computation
30: of likelione (using inter/intrapolation if mle = 0) in order to
31: get same likelihood as if mle=1.
32: Some cleaning of code and comments added.
33:
34: 2009-10-29 brouard <brouard@ined.fr>
35:
36: * imach.c (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
37:
38: 2006-06-30 Brouard Nicolas <brouard@localhost>
39:
40: * imach.c (Module): Clarifications on computing e.j
41:
42: 2006-04-28 Brouard Nicolas <brouard@localhost>
43:
44: * imach.c (Module): Yes the sum of survivors was wrong since
45: imach-114 because nhstepm was no more computed in the age
46: loop. Now we define nhstepma in the age loop.
47: (Module): In order to speed up (in case of numerous covariates) we
48: compute health expectancies (without variances) in a first step
49: and then all the health expectancies with variances or standard
50: deviation (needs data from the Hessian matrices) which slows the
51: computation.
52: In the future we should be able to stop the program is only health
53: expectancies and graph are needed without standard deviations.
54:
55: 2006-04-04 LIEVRE Agnes <lievre@PC_00697.ined.fr>
56:
57: * imach.iss (Module):
58:
59: * imach.c (Module): bugs in evsi + population file added in htm output file
60:
61: 2006-03-22 LIEVRE Agnes <lievre@PC_00697.ined.fr>
62:
63: * imach.c (Module):
64:
65: 2006-03-20 Brouard Nicolas <brouard@localhost>
66:
67: * imach.c (Module): <title> changed, corresponds to .htm file
68: name. <head> headers where missing.
69:
70: * Makefile (Module): 0.98g
71:
72: * imach.c (Module): Weights can have a decimal point as for
73: English (a comma might work with a correct LC_NUMERIC environment,
74: otherwise the weight is truncated).
75: Modification of warning when the covariates values are not 0 or
76: 1.
77: Version 0.98g
78:
79: 2006-03-16 LIEVRE Agnes <lievre@PC_00697.ined.fr>
80:
81: * imach.c (Module): Comments concerning covariates added
82:
83: * noreg-setup.iss (Module): Version 0.98f
84:
85: * Makefile (Module): Version 0.98f make _windows=1 imachcyg-setup
86:
87: * imach.c (Module): refinements in the computation of lli if
88: status=-2 in order to have more reliable computation if stepm is
89: not 1 month. Version 0.98f
90:
91: 2006-03-15 Brouard Nicolas <brouard@localhost>
92:
93: * imach.c (Module): Bug if status = -2, the loglikelihood was
94: computed as likelihood omitting the logarithm. Version O.98e
95:
96: 2006-03-14 Brouard Nicolas <brouard@localhost>
97:
98: * imach.c (Module): varevsij Comments added explaining the second
99: table of variances if popbased=1 .
100: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
101: (Module): Function pstamp added
102: (Module): Version 0.98d
103:
104: 2006-03-06 Brouard Nicolas <brouard@localhost>
105:
106: * imach.c (Module): Variance-covariance wrong links and
107: varian-covariance of ej. is needed (Saito).
108:
109: 2006-02-27 Brouard Nicolas <brouard@localhost>
110:
111: * imach.c (Module): One freematrix added in mlikeli!
112:
113: 2006-02-26 Brouard Nicolas <brouard@localhost>
114:
115: * imach.c (Module): Some improvements in processing parameter
116: filename with strsep.
117:
118: 2006-02-24 Brouard Nicolas <brouard@localhost>
119:
120: * imach.c (Module): Memory leaks checks with valgrind and:
121: datafile was not closed, some imatrix were not freed and on matrix
122: allocation too.
123:
124: 2006-02-09 LIEVRE Agnes <lievre@PC_00697.ined.fr>
125:
126: * strsep.c (Module): Inclusion because not in mingw32!!!
127:
128: * setup.iss.in (Module): IMACHSETUPVERSION added
129:
130: * noreg-setup.iss (Module): added to cvs tree
131:
132: * Makefile (Module): IMACHSETUPVERSION added
133:
134: 2006-01-30 BROUARD Nicolas <brouard@localhost>
135:
136: * Makefile (Module, Module): Latest Makefile
137:
138: * imach.c (Module): Back to gnuplot.exe instead of wgnuplot.exe
139:
140: 2006-01-27 BROUARD Nicolas <brouard@localhost>
141:
142: * noreg-setup.iss (Module): ADDED
143:
144: 2006-01-24 BROUARD Nicolas <brouard@localhost>
145:
146: * imach.c (Module): Comments (lines starting with a #) are allowed in data.
147:
148: 2005-10-25 Nicolas Brouard <brouard@dhcp171.recherche.ined.fr>
149:
150: * Makefile (Module): Added Imach98.pmsp and howto crete a dmg
151:
152: * createdmg.sh (Module): Creates a dmg disk for MaC OS/X
153:
154: 2005-09-30 <lievre@PC_00595>
155:
156: * imach.c (Module): sump fixed, loop imx fixed, and simplifications.
157: (Module): If the status is missing at the last wave but we know
158: that the person is alive, then we can code his/her status as -2
159: (instead of missing=-1 in earlier versions) and his/her
160: contributions to the likelihood is 1 - Prob of dying from last
161: health status (= 1-p13= p11+p12 in the easiest case of somebody in
162: the healthy state at last known wave). Version is 0.98
163:
164: 2004-05-20 Brouard Nicolas <brouard@localhost>
165:
166: * imach.c (Repository):
167: Agnes added a direct estimation of mortality (without the need of
168: computing period prevalence and differential mortality). Thus here
169: is version 0.97a which has been distributed to some people at
170: REVES 16 in Brugge using an Inno setup.exe for PCs. Estimates of
171: mortality using covariates is not done today. Estimating direct
172: mortality is a very different process because it doesn't need
173: interpolation because it is easy to get the lx from the force of
174: the mortality mux in the simplest case as for a Gompertz (log mux
175: = a + b*x . But we have been able to incorporate the new code
176: within former imach program (0.96d) without deteriorating too much
177: the understanding of the program.
178:
179: 2003-06-25 <brouard@BROUARD>
180:
181: * imach.c (Module): On windows (cygwin) function asctime_r doesn't
182: exist so I changed back to asctime which exists.
183: (Module): Version 0.96b
184:
185: 2003-06-24 <brouard@BROUARD>
186:
187: * imach.c (Module): Some bugs corrected for windows. Also, when
188: mle=-1 a template is output in file "or"mypar.txt with the design
189: of the covariance matrix to be input.
190:
191: 2003-06-25 Brouard Nicolas <brouard@homebrou>
192:
193: * imach.c (Repository): Duplicated warning errors corrected.
194: (Repository): Elapsed time after each iteration is now output. It
195: helps to forecast when convergence will be reached. Elapsed time
196: is stamped in powell. We created a new html file for the graphs
197: concerning matrix of covariance. It has extension -cov.htm.
198:
199: 2003-06-23 Brouard Nicolas <brouard@homebrou>
200:
201: * 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.
202:
203: 2003-06-18 Brouard Nicolas <brouard@homebrou>
204:
205: * imach.c (Repository): Suppress abusive calls to append to file html (solution was to put fichtm as a global variable. Version 0.96
206:
207: 2003-06-17 Brouard Nicolas <brouard@homebrou>
208:
209:
210: * imach.c (Repository): Check when date of death was earlier that
211: current date of interview. It may happen when the death was just
212: prior to the death. In this case, dh was negative and likelihood
213: was wrong (infinity). We still send an "Error" but patch by
214: assuming that the date of death was just one stepm after the
215: interview.
216: (Repository): Because some people have very long ID (first column)
217: we changed int to long in num[] and we added a new lvector for
218: memory allocation. But we also truncated to 8 characters (left
219: truncation)
220:
221: (Repository): No more line truncation errors.
222:
223: 2003-06-14 Brouard Nicolas <brouard@homebrou>
224:
225: * imach.c (Repository): Add of a routine likelione (likelihood
226: only once) which prints on a text file
227: (ilk) the contributions to the likelihood for each
228: individual/wave.
229:
230: 2003-06-13 Brouard Nicolas <brouard@homebrou>
231: * imach.c (Repository): Replace "freqsummary" at a correct
232: place. It differs from routine "prevalence" which may be called
233: many times. Probs is memory consuming and must be used with
234: parcimony.
235:
236: 2003-06-17 <brouard@BROUARD>
237:
238: * timeval.h (Module): Added included file to make use of
239: gettimeofday working on win32 with cygwin.
240:
241: 2003-05-16 Brouard Nicolas <brouard@brouard>
242: * imach.c (Module): Insert of a warning if the delay between two
243: waves is negative.
244:
245: * imach.c (Module):
246:
247: 2003-05-03 <brouard@BROUARD>
248:
249: * imach.c: Some cleaning
250:
251: 2003-03-28 <brouard@BROUARD>
252: * imach.c (Module): In version up to 0.92 likelihood was computed
253: as if date of death was unknown. Death was treated as any other
254: health state: the date of the interview describes the actual state
255: and not the date of a change in health state. The former idea was
256: to consider that at each interview the state was recorded
257: (healthy, disable or death) and IMaCh was corrected; but when we
258: introduced the exact date of death then we should have modified
259: the contribution of an exact death to the likelihood. This new
260: contribution is smaller and very dependent of the step unit
261: stepm. It is no more the probability to die between last interview
262: and month of death but the probability to survive from last
263: interview up to one month before death multiplied by the
264: probability to die within a month. Thanks to Chris
265: Jackson for correcting this bug. Former versions increased
266: mortality artificially. The bad side is that we add another loop
267:
268:
269: which slows down the processing. The difference can be up to 10%
270:
271:
272: lower mortality.
273:
274:
275:
276:
277:
278: 2002-05-30 <brouard@BROUARD>
279:
280:
281:
282:
283:
284: *
285:
286:
287:
288:
289:
290: * imach.c (Module): Add correlation matrix of one-step
291:
292:
293: probabilities and covariance matrix
294:
295:
296:
297:
298:
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