440a0e460b
Storage of results: use subfolder
...
Ref. #1758
2021-07-23 14:12:17 +02:00
80cad32068
Particle filter: enable passing of option values and prepare for other samplers
2021-07-21 23:00:32 +02:00
02072dde39
Add Occbin routines
...
Syntax is not yet finalized (see preprocessor#68).
Documentation still to be done.
Ref. #569
2021-07-16 17:20:11 +02:00
f7a232061a
Remove unused options_-output
...
Prevents accidentally changing something
2021-06-22 18:17:11 +02:00
766fff88f6
Use secure URL for link to GNU licenses
2021-06-09 17:35:05 +02:00
5c3b0ebc95
better handle newrat options, avoiding to treat newratflag in two places
2021-02-13 16:44:50 +01:00
b9a6462462
bug fix in trapping option newratflag
2021-02-12 18:54:07 +01:00
96b1ca3bee
mr_hessian.m: honor Save_files option
2021-01-27 21:38:03 +01:00
09f992d584
Merge branch 'moments_varendo' of git.dynare.org:JohannesPfeifer/dynare
2021-01-25 18:18:46 +01:00
992e375bfa
Merge branch 'use_mh_covariance' of git.dynare.org:JohannesPfeifer/dynare
2021-01-25 18:16:23 +01:00
c27c4fd932
moments_varendo: delete stale files from previous runs
2021-01-25 15:55:10 +01:00
de16e9c00a
dynare_estimation_1: restore exiting with mode_compute=0 and order=1
2021-01-22 21:19:07 +01:00
417f038cc5
Particle filters: do not compute Kalman smoother
...
Related to https://git.dynare.org/Dynare/dynare/-/issues/1757
2021-01-22 21:19:07 +01:00
731a168f69
Make mh_tune_jscale work with use_mh_covariance_matrix
2021-01-22 15:01:23 +01:00
ea03868a6b
mode_compute=0: run dyn_forecast after smoother when forecast > 0
2021-01-10 19:46:18 +01:00
da3943beba
Make sure output is saved in dname-folder
...
That's where other functions are looking for them
2020-12-16 22:41:55 +01:00
6a6271225c
Merge branch 'minimize_objective' into 'master'
...
Minimize objective
See merge request Dynare/dynare!1737
2020-07-07 09:35:39 +00:00
e92054e16b
Move definition of prior distribution names to separate function to make it useable in other contexts.
2020-06-26 12:07:16 +02:00
4a9f9263df
Make newrat and gmhmaxlik independent of varargin inputs
...
Makes the optimizers useable in other contexts where varargin is different
2020-06-23 15:04:11 +02:00
312d053ae1
Use LaTeX labels in figures
2020-06-17 21:48:25 +02:00
d01846e5be
Fix incorrect test for options_.posterior_sampler_options.posterior_sampling_method
...
Thanks to checkcode().
2020-03-10 14:13:49 +01:00
79cca46dd6
trap error when sub_draws is set to zero
2020-01-24 10:22:24 +01:00
49dc997073
Global reindentation of MATLAB code (excluding submodules)
...
Also convert to Unix end-of-lines, and remove trailing whitespaces.
2019-12-20 16:30:27 +01:00
Stéphane Adjemian (Charybdis)
227b2661cb
Do not trap order>2 in estimation.
2019-12-20 11:31:56 +01:00
a327e2c9b8
Merge branch 'tex_subfolder' into 'master'
...
Move remaining LaTeX-file to subfolder
Closes #1655
See merge request Dynare/dynare!1667
2019-11-21 14:15:54 +00:00
8e5fd8dc4e
Move creation of subfolder outside of TeX-check
...
Required if we only want the figure
2019-11-21 11:58:16 +01:00
0075776716
Laplace approximation: filter out non-positive definite Hessians
...
Closes https://git.dynare.org/Dynare/dynare/issues/1659
2019-11-21 11:32:13 +01:00
a59534a284
Move remaining figures and related TeX output from main folder to graphs
2019-11-21 10:14:48 +01:00
89a3e94cbf
Use true/false instead of 1/0 for boolean options
...
This is more elegant, and makes it easier to distinguish them from integer
options.
Also simplify test expressions for these boolean options.
2019-03-19 15:21:16 +01:00
7a75872f72
Modernization: use tilde (~) syntax for ignored output arguments
2018-11-13 18:02:09 +01:00
f53be721c1
Added new option mh_tune_jscale.
...
Works only with the Random Walk Metropolis Hastings algorithm.
Closes #1598
2018-05-16 16:37:22 +02:00
127a543b65
prior bounds: make sure that lower and upper bounds only affect mode-finding
...
99dbc8c74d
2018-02-01 10:16:47 +01:00
7be8f10e0e
Use cells of strings instead of char arrays.
2018-01-09 22:30:01 +01:00
ce7cfccddb
Cosmetic changes.
2017-10-10 10:14:59 +02:00
da063b2e18
bug fix in optimizer 5 [newrat] for dsge_var_likelihood objective function. Fixes #1494
2017-10-09 17:53:19 +02:00
818a07a4ef
Fix get_posterior_parameters
...
Get rid of setting variables from the base workspace in the function, making the function read-only (as the name suggests)
2017-10-05 09:54:08 +02:00
16a3e11034
Transform hard-coded plot tolerance into an undocumented option
2017-08-16 19:51:57 +02:00
4cd259baa6
Fix plotting of measurement errors in non-MCMC case
...
Number of plots was wrongly computed
2017-08-16 19:51:57 +02:00
5417b27ac7
Fixed indentation of matlab files.
2017-05-16 15:10:20 +02:00
88e1701289
Removed useless commas and semicolons.
2017-05-16 13:24:46 +02:00
00299a92c1
Make sure options_.first_obs is properly set
2017-04-05 11:44:27 +02:00
732aa8c30c
Merge remote-tracking branch 'rattoma/slice'
2017-03-24 16:19:04 +01:00
e5f1d20221
change dyn_figure to take a nodisplay and graph_format arguments as opposed to options_. closes #1415
2017-03-23 17:59:05 +01:00
2aa85f978a
change dyn_figure to take a nodisplay argument as opposed to options_. #1415
2017-03-23 14:30:16 +01:00
99dbc8c74d
bug fix with redundant call to prior_bounds
2017-03-22 23:08:18 +01:00
1e3e2f70a9
set posterior sampler options when load_mh_file
2017-03-18 00:06:15 +01:00
cc34e16458
Set temporarily mode_compute and mh_replic to zero if estim_params_ is empty.
2017-03-16 15:45:08 +01:00
9fbef0c107
Removed penalty_hessian routine.
...
+ Code factorization.
+ Added an option for using the penalized objective when computing numerically
the hessian at the mode.
Previous behaviour (introduced with penalty_hessian routine) was to compute the
hessian matrix at the mode with the penalized objective function (instead of
the original objective function). This behaviour hides problematic situations,
where the computed hessian (using the original objective) would not be full
rank. For instance, if the estimation ends up with a parameter on (or very
close to) the bounds of its possible values (which is often not a desirable
outcome), the estimated posterior variance would be zero for this
parameter (with the original objective) because the hessian is not finite in
this direction, while the posterior variance would be positive if the penalized
objective is used instead. But this estimate would not be reliable by
construction of the penalty which is quite ad-hoc (more fundamentally I do not
think that there exists any rational for approximating the covariance matrix
with the inverse of the hessian matrix if the mode is on the border of the set
of possible values).
This commit restore the behaviour previous to 2446ab02ba
2017-02-08 13:07:25 +00:00
7b4fc9ec4a
Moving loading of MCMC_jumping_covariance to after display of standard errors and computation of Laplace approximation
...
Closes #1255
2016-12-27 12:07:22 +01:00
352786c2b4
load_mh_file: recompute results by default and load on request
...
Prevents general overwriting of results in oo_ and having stale results
2016-12-19 19:37:34 +01:00
Johannes Pfeifer
Marco Ratto
Sébastien Villemot
Stéphane Adjemian(Charybdis)
Houtan Bastani
Marco Ratto