Johannes Pfeifer
c2e6feeb75
Deletion of unused definition of H_is_diagonal
...
Cosmetic fix
2013-09-04 11:11:29 +02:00
Johannes Pfeifer
e1a0fd0524
Bugfix for conditional variance decomposition
...
sigma_e_is_diagonal was erroneously set to 0 if the covariance matrix was diagonal, but contained a 0 on the main diagonal.
2013-09-04 11:11:14 +02:00
Johannes Pfeifer
01c3724371
Deal with pathological case where nobs=1
...
In this case the plots crash due to an invalid xlim. Also adds markers if only one point is plotted
2013-08-15 16:35:01 +02:00
Sébastien Villemot
b4f8532bea
Merge pull request #441 from JohannesPfeifer/mode_check
...
Bugfix for mode_check and display of parameters out of bounds
2013-08-12 03:31:50 -07:00
Stéphane Adjemian (Charybdis)
184c403375
Replaced disp(' ') by skipline().
2013-07-10 16:16:32 +02:00
Stéphane Adjemian (Charybdis)
964b7580d8
Cosmetic change.
2013-07-10 16:15:28 +02:00
Stéphane Adjemian (Charybdis)
dfc51d097a
Cosmetic change. Replaced warning(...) by disp(...).
2013-07-10 16:09:31 +02:00
Stéphane Adjemian (Charybdis)
2fae989131
Rewrote the portion of code requesting a user input when non linear filters are used with a gradient based optimization routine.
...
Closes issues #442 and #443 .
2013-07-10 16:03:49 +02:00
Johannes Pfeifer
aeda8d17ff
Add warning if gradient-based optimizer is used with particle-filtering and offer user choice to change mode-finder
2013-07-10 12:08:59 +02:00
Johannes Pfeifer
41f6c2ff66
Display source of non-positive definite Hessian if parameters at prior bound
2013-07-06 00:26:23 +02:00
Johannes Pfeifer
d9579418fd
Bugfix for correct display of names of parameters outside of bounds
2013-07-06 00:20:07 +02:00
Stéphane Adjemian (Charybdis)
08feada8b0
Fixed bug.
2013-06-24 10:24:55 +02:00
Stéphane Adjemian (Charybdis)
340bcc2739
Added an option for setting the number of points where the posterior kernel is evaluated when using mode_check. Reorganized the options_ global structure wrt mode_check.
2013-06-20 17:19:44 +02:00
Frédéric Karamé
1bbfd90498
Added online auxiliary filter as options_.mode_compute=10.
2013-06-18 16:18:55 +02:00
Stéphane Adjemian (Charybdis)
c08c34223d
Fixed typo and bug in MCMC diagnostics.
2013-06-13 17:14:04 +02:00
Johannes Pfeifer
e8ed4c1f67
Bunch of cosmetic changes
2013-06-13 14:24:17 +02:00
Stéphane Adjemian
fc83427185
Merge pull request #393 from JohannesPfeifer/display_tables
...
Add function for displaying estimation results
2013-05-14 03:18:19 -07:00
Johannes Pfeifer
d9fb19e479
Make error message if mode violated prior more explicit
2013-05-12 13:27:55 +02:00
Johannes Pfeifer
7aea75f348
Add function for displaying estimation results
...
The commit moves the creation of results tables, both the ones displayed
and printed to LaTeX, from dynare_estimation_1.m to a separate function.
This factorization later allows to use this function to display results
from other estimation routines. At the same time, it removes the display
of t-statistics for posterior estimates ( #339 )
2013-05-10 18:58:08 +02:00
Johannes Pfeifer
58b6c3122f
Cosmetic Changes to Graphs
2013-04-26 19:57:36 +02:00
Johannes Pfeifer
7518072e77
Fix several bugs related to estimated measurement errors
...
1. The first call to set_prior overwrote the first column of
estim_params_.var_endo storing the position of the variable with
measurement error in M_.endo_names with the position in
options_.var_obs. All subsequent calls to set_prior then lead to
crashes.
2. At the same time, for correlations of ME, the first column of
estim_params_.corrn still stored the position of the variable with
measurement error in M_.endo_names. But subsequent calls to it were done
as if it stored the position in options_.var_obs
I introduced two new variables in estim_params_ storing the respective
positions in var_obs so as to not necessitate changes in the
preprocessors.
3. For cases of calibrated measurement error correlations, the
covariance matrix was not updated.
4. Fixing a lot of smaller bugs related to measurement errors, including
some copy and paste errors
-
2013-03-21 20:51:12 +01:00
Johannes Pfeifer
6390830b4d
Store MCMC information recorded in record in oo_
...
Closes issue #315 (https://github.com/DynareTeam/dynare/issues/315 )
2013-03-17 22:49:28 +01:00
Sébastien Villemot
5c8efaae6c
More explicit error message when posterior mode file is outdated
2013-03-08 14:31:15 +01:00
Sébastien Villemot
5c88338965
Fix mode_compute=3 under Octave
2013-01-07 15:40:48 +01:00
Sébastien Villemot
8c7e3d46ed
Document when mode_compute={1,3,7} are available, and error with an informative message if not
2013-01-07 15:40:45 +01:00
Sébastien Villemot
c06bd0ae75
Support nograph option for posterior distribution plots
2012-11-29 14:52:33 +01:00
Michel Juillard
c722f53955
fixed bug with options mode_compute=0,mode_file=....
...
added tests for mode_compute=4 and mode_compute=6. They increase computing time of the
test suite by about 30 minutes.
2012-10-06 16:51:42 +02:00
Stéphane Adjemian (Charybdis)
e81f9d48ac
Improved display of mode_compute=6 optimization algorithm. Fixed bugs. Changed the options.
2012-09-29 00:08:05 +02:00
Stéphane Adjemian (Charybdis)
3744061112
Save the names of the estimated parameters in <MODEL_NAME>_mode.mat when mode_compute is equal to six (stochastic optimization).
2012-09-29 00:08:05 +02:00
Marco Ratto
e519b04713
bug fix: when nograph=1, SmoothedShocks were not saved.
2012-08-27 17:45:47 +02:00
Marco Ratto
40e1e60fd1
1) changed options of optimizer n. 1 with analytic derivs;
...
2) force no analytic derivative with mode_check;
2012-07-05 10:22:36 +02:00
Stéphane Adjemian (Charybdis)
b907e82e7b
Removed NaNs in M_.Correlation_matrix if they are the consequence of variances (of structural innovations) calibrated to zero.
2012-06-14 15:32:14 +02:00
Marco Ratto
2fecf9946b
1) Extended optimizer = 5 for analytic derivatives;
...
2) Start adapting identification routines to allow computation of analytic asymptotic Hessian with KF routines
2012-06-08 14:23:18 +02:00
Stéphane Adjemian (Charybdis)
0da05ae29d
Removed globals from set_all_parameters routine.
2012-06-07 15:13:39 +02:00
Stéphane Adjemian (Charybdis)
e12748329b
Fixed bug reported by Johannes ( http://www.dynare.org/pipermail/dev/2012-May/002015.html ), see also trac#257.
2012-06-06 18:25:08 +02:00
Stéphane Adjemian (Charybdis)
fac9528014
Do not compute smoothed variables or forecasts if the model is
...
estimated with a particle filter (the non linear smoother is not yet
implemented).
2012-06-06 18:25:08 +02:00
Stéphane Adjemian (Charybdis)
ef1146f5a3
Fixed bug trac#202. Define the sigma_e_is_diagonal flag when no shocks block is declared in the mod file.
2012-06-06 18:25:08 +02:00
Sébastien Villemot
9bea42f411
order=2 in estimation command now triggers particle filter
2012-06-06 17:08:53 +02:00
Marco Ratto
4480f5b494
Fixes around latex output: use of longtable for parameter estimates and various fixes for posterior analysis, also for parallel execution.
2012-05-30 11:22:59 +02:00
Johannes Pfeifer
c0ffcf7fd8
Fix problem with smoother where original data series was not plotted
2012-05-11 19:58:16 +02:00
Marco Ratto
da9ec0f187
Estimation with analytic scores and hessian;
...
This includes re-setting the list of output arguments in objective functions
Added test function
2012-04-29 21:18:33 +02:00
Michel Juillard
e692185c6b
storing oo_.prior.mean oo_.prior.variance
...
oo_.posterior.optimization.mode oo_.posterior.optimization.variance
oo_.posterior.metropolis.mean oo_.posterior.metropolis.variance as
aggregate arrays in addition to previous storage variable by variable
2012-04-21 21:28:03 +02:00
Michel Juillard
973302e42e
fixing bugs in smoother
2012-04-18 21:15:18 +02:00
Stéphane Adjemian (Charybdis)
9cc4661942
Fixed bug in initial_estimation_checks + Cosmetic changes.
2012-03-30 11:12:59 +02:00
Michel Juillard
5d4d318bc7
fixing bug in estimation by maximum likelihood
2012-03-15 15:33:02 +01:00
Stéphane Adjemian (Charybdis)
c5fa0b82a9
Merge remote-tracking branch 'marco/master'
2012-03-14 12:49:56 +01:00
Stéphane Adjemian (Charybdis)
cf9e8a6714
Bug fix. When the estimation command is used, do not issue an error message if options_.order>1. In this
...
case reset the options_.order equal to one (except if options_.particle.status==1).
2012-03-05 14:56:02 +01:00
Stéphane Adjemian (Charybdis)
59d3d30c90
Fixed typo. Do not use the non linear filter automatically when order==2, otherwise the non linear filters are used if the estimation command is used after a stoch_simul command with order==2. The non linear filter is used only if options_.particle.status is explicitely set to 1 (before the estimation command). A new estimation option is needed...
2012-03-05 14:21:52 +01:00
Stéphane Adjemian (Charybdis)
e1a5e0de53
Objective function needs to be passed as a string to CMAES optimization routine.
2012-03-04 22:13:00 +01:00
Stéphane Adjemian (Charybdis)
6a8095303e
Added particle filter flag.
2012-03-04 22:12:59 +01:00