File: change_log.html

package info (click to toggle)
dynare 4.5.7-1
  • links: PTS, VCS
  • area: main
  • in suites: buster
  • size: 49,408 kB
  • sloc: cpp: 84,998; ansic: 29,058; pascal: 13,843; sh: 4,833; objc: 4,236; yacc: 3,622; makefile: 2,278; lex: 1,541; python: 236; lisp: 69; xml: 8
file content (280 lines) | stat: -rw-r--r-- 7,871 bytes parent folder | download | duplicates (5)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
<HTML>
<TITLE>
Dynare++ Change Log
</TITLE>
<!-- $Header$ -->
<BODY>
<TABLE CELLSPACING=2 ALIGN="CENTER" BORDER=1>
<TR>
<TD BGCOLOR="#d0d0d0" WIDTH="85"> <b>Revision</b>    </TD>
<TD BGCOLOR="#d0d0d0" WIDTH="85"> <b>Version</b></TD>
<TD BGCOLOR="#d0d0d0" WIDTH="80"> <b>Date</b>   </TD>
<TD BGCOLOR="#d0d0d0" WIDTH="600"> <b>Description of changes</b></TD>
</TR>

<TR>
<TD>
<TD>1.3.7
<TD>2008/01/15
<TD>

<TR><TD><TD><TD> <TD> Corrected a serious bug in centralizing a
decision rule. This bug implies that all results based on simulations
of the decision rule were wrong. However results based on stochastic
fix points were correct. Thanks to Wouter J. den Haan and Joris de Wind!

<TR><TD><TD><TD> <TD> Added options --centralize and --no-centralize.

<TR><TD><TD><TD> <TD> Corrected an error of a wrong
variance-covariance matrix in real-time simulations (thanks to Pawel
Zabzcyk).

<TR><TD><TD><TD> <TD> Corrected a bug of integer overflow in refined
faa Di Bruno formula if one of refinements is empty. This bug appeared
when solving models without forward looking variables.

<TR><TD><TD><TD> <TD> Corrected a bug in the Sylvester equation
formerly working only for models with forward looking variables.

<TR><TD><TD><TD> <TD> Corrected a bug in global check printout.

<TR><TD><TD><TD> <TD> Added generating a dump file.

<TR><TD><TD><TD> <TD> Fixed a bug of forgetting repeated assignments
(for example in parameter settings and initval).

<TR><TD><TD><TD> <TD> Added a diff operator to the parser.

<TR>
<TD>1539
<TD>1.3.6
<TD>2008/01/03
<TD>

<TR><TD><TD><TD> <TD> Corrected a bug of segmentation faults for long
names and path names. 

<TR><TD><TD><TD> <TD> Changed a way how random numbers are
generated. Dynare++ uses a separate instance of Mersenne twister for
each simulation, this corrects a flaw of additional randomness caused
by operating system scheduler. This also corrects a strange behaviour
of random generator on Windows, where each simulation was getting the
same sequence of random numbers.

<TR><TD><TD><TD> <TD> Added calculation of conditional distributions
controlled by --condper and --condsim.

<TR><TD><TD><TD> <TD> Dropped creating unfoled version of decision
rule at the end. This might consume a lot of memory. However,
simulations might be slower for some models.

<TR>
<TD>1368
<TD>1.3.5
<TD>2007/07/11
<TD>

<TR><TD><TD><TD> <TD> Corrected a bug of useless storing all derivative
indices in a parser. This consumed a lot of memory for large models.

<TR><TD><TD><TD> <TD> Added an option <tt>--ss-tol</tt> controlling a
tolerance used for convergence of a non-linear solver.

<TR><TD><TD><TD> <TD> Corrected buggy interaction of optimal policy
and forward looking variables with more than one period.

<TR><TD><TD><TD> <TD> Variance matrices can be positive
semidefinite. This corrects a bug of throwing an error if estimating
approximation errors on ellipse of the state space with a
deterministic variable.

<TR><TD><TD><TD> <TD> Implemented simulations with statistics
calculated in real-time. Options <tt>--rtsim</tt> and <tt>--rtper</tt>.

<TR>
<TD>1282
<TD>1.3.4
<TD>2007/05/15
<TD>

<TR><TD><TD><TD> <TD>Corrected a bug of wrong representation of NaN in generated M-files.

<TR><TD><TD><TD> <TD>Corrected a bug of occassionaly wrong evaluation of higher order derivatives of integer powers.

<TR><TD><TD><TD> <TD>Implemented automatic handling of terms involving multiple leads.

<TR><TD><TD><TD> <TD>Corrected a bug in the numerical integration, i.e. checking of the precision of the solution.

<TR>
<TD>1090
<TD>1.3.3
<TD>2006/11/20
<TD>

<TR><TD><TD><TD> <TD>Corrected a bug of non-registering an auxiliary variable in initval assignments.

<TR>
<TD>988
<TD>1.3.2
<TD>2006/10/11
<TD>

<TR><TD><TD><TD> <TD>Corrected a few not-serious bugs: segfault on
some exception, error in parsing large files, error in parsing
matrices with comments, a bug in dynare_simul.m

<TR><TD><TD><TD> <TD>Added posibility to specify a list of shocks for
which IRFs are calculated

<TR><TD><TD><TD> <TD>Added --order command line switch

<TR><TD><TD><TD> <TD>Added writing two Matlab files for steady state
calcs

<TR><TD><TD><TD> <TD>Implemented optimal policy using keyword
planner_objective and planner_discount

<TR><TD><TD><TD> <TD>Implemented an R interface to Dynare++ algorithms
(Tamas Papp)

<TR><TD><TD><TD> <TD>Highlevel code reengineered to allow for
different model inputs

<TR>
<TD>799
<TD>1.3.1
<TD>2006/06/13
<TD>

<TR><TD><TD><TD> <TD>Corrected few bugs: in error functions, in linear algebra module.

<TR><TD><TD><TD> <TD>Updated dynare_simul.

<TR><TD><TD><TD> <TD>Updated the tutorial.

<TR><TD><TD><TD> <TD>Corrected an error in summing up tensors where
setting up the decision rule derivatives. Thanks to Michel
Juillard. The previous version was making deterministic effects of
future volatility smaller than they should be.

<TR>
<TD>766
<TD>1.3.0
<TD>2006/05/22
<TD>

<TR><TD><TD><TD> <TD>The non-linear solver replaced with a new one.

<TR><TD><TD><TD> <TD>The parser and derivator replaced with a new
code. Now it is possible to put expressions in parameters and initval
sections.

<TR>
<TD>752
<TD>1.2.2
<TD>2006/05/22
<TD>

<TR><TD><TD><TD> <TD>Added an option triggering/suppressing IRF calcs..

<TR><TD><TD><TD> <TD>Newton algortihm is now used for fix-point calculations.

<TR><TD><TD><TD> <TD> Vertical narrowing of tensors in Faa Di Bruno
formula to avoid multiplication with zeros..

<TR>
<TD>436
<TD>1.2.1
<TD>2005/08/17
<TD>

<TR><TD><TD><TD> <TD>Faa Di Bruno for sparse matrices optimized. The
implementation now accommodates vertical refinement of function stack
in order to fit a corresponding slice to available memory. In
addition, zero slices are identified. For some problems, this implies
significant speedup.

<TR><TD><TD><TD> <TD>Analytic derivator speedup.

<TR><TD><TD><TD> <TD>Corrected a bug in the threading code. The bug
stayed concealed in Linux 2.4.* kernels, and exhibited in Linux 2.6.*,
which has a different scheduling. This correction also allows using
detached threads on Windows.

<TR>
<TD>410
<TD>1.2
<TD>2005/07/29
<TD>

<TR><TD><TD><TD> <TD>Added Dynare++ tutorial.

<TR><TD><TD><TD> <TD>Changed and enriched contents of MAT-4 output
file.

<TR><TD><TD><TD> <TD>Corrected a bug of wrong variable indexation
resulting in an exception. The error occurred if a variable appeared
at time t-1 or t+1 and not at t.

<TR><TD><TD><TD> <TD>Added Matlab interface, which allows simulation
of a decision rule in Matlab.

<TR><TD><TD><TD> <TD>Got rid of Matrix Template Library.

<TR><TD><TD><TD> <TD>Added checking of model residuals by the
numerical integration. Three methods: checking along simulation path,
checking along shocks, and on ellipse of states.

<TR><TD><TD><TD> <TD>Corrected a bug in calculation of higher moments
of Normal dist.

<TR><TD><TD><TD> <TD>Corrected a bug of wrong drawing from Normal dist
with non-zero covariances.

<TR><TD><TD><TD>
<TD>Added numerical integration module. Product and Smolyak
quadratures over Gauss-Hermite and Gauss-Legendre, and quasi Monte
Carlo.

<TR>
<TD>152
<TD>1.1
<TD>2005/04/22
<TD>

<TR><TD><TD><TD>
<TD>Added a calculation of approximation at a stochastic steady state
(still experimental).

<TR><TD><TD><TD>
<TD>Corrected a bug in Cholesky decomposition of variance-covariance
matrix with off-diagonal elements.

<TR>
<TD>89
<TD>1.01
<TD>2005/02/23
<TD>

<TR><TD><TD><TD>
<TD>Added version printout.

<TR><TD><TD><TD>
<TD>Corrected the bug of multithreading support for P4 HT processors running on Win32.

<TR><TD><TD><TD>
<TD>Enhanced Kronecker product code resulting in approx. 20% speedup.

<TR><TD><TD><TD>
<TD>Implemented vertical stack container refinement, and another
method for sparse folded Faa Di Bruno (both not used yet).

<TR>
<TD>5
<TD>1.0
<TD>2005/02/23
<TD>The first released version.

</TABLE>
</BODY>
</HTML>