Commit Graph

5534 Commits (ebc7d6f67aaa16d36c846f5582e6fc997ace3174)

Author SHA1 Message Date
Johannes Pfeifer ebc7d6f67a Add comment why use of old correlation matrix from previous draw is correct and revert change
Due to only using the diagonal of Sigma_e and the correlation matrix having ones on the diagonal, the diagonal entries of the covariance matrix are correctly built from recent draw. Later, when using the new draw for the correlations, only the correctly updated diagonal entries of Sigma_e are used.
2013-03-21 20:49:33 +01:00
Johannes Pfeifer 74ef1aa7ed Fixing bug introduced in ae82c284dc 2013-03-19 20:48:09 +01:00
Johannes Pfeifer 156397ec31 Fix bug in ML forecasting if observables are not included 2013-03-19 19:04:03 +01:00
Sébastien Villemot 3c5aab9e20 Merge pull request #327 from rattoma/master
Identification and sensitivity fixes
2013-03-19 10:52:05 -07:00
Marco Ratto 597d30d0bf bug fix in call to priordens; 2013-03-19 17:07:14 +01:00
Marco Ratto 92e4e73765 Trap the cases where the state space matrices are not stored for memory allocation issues
[to be fixed later by breaking the big matrix stored ...]
2013-03-19 17:06:33 +01:00
Marco Ratto b0ce2dacd9 1) better trace all cases when the model does not solve, classifying cases according to info output of dynare_resolve;
2) trap the case of memory allocation limit for too large model undergoing to reduced form mapping
2013-03-19 17:04:33 +01:00
Marco Ratto d6993cecc4 1) trap the case where the model does not solve for any parameter values in the MC sample;
2) fixed initialization of alpha2_stab;
2013-03-19 17:02:31 +01:00
Marco Ratto 22e0b9cc3d 1) fixed bugs when no estimated parameters are declared;
2) fixed bug for the ML base (bayestopt_.pshape==0);
3) make more transparent the error when the model does not solve at prior_mode or  prior_mean;
4) when prior_mean or prior_mode fail, try 50 times with randomly generated samples from the prior before interrupting identification.

This commit fixes issues related to fabiac forum request and subsequent discussion with Johannes.

http://www.dynare.org/phpBB3/viewtopic.php?f=1&t=4402&p=10771&hilit=identification&sid=64c6f9d987a2641e79dd5722137eb483#p10771
2013-03-19 16:56:10 +01:00
Houtan Bastani 6f2ff28d1b reporting: @table: only print given date range 2013-03-19 14:35:05 +01:00
Houtan Bastani 49586d0b54 reporting: add centering 2013-03-19 14:34:56 +01:00
Houtan Bastani e02e06b03f reporting: @table: simplify code 2013-03-19 12:51:02 +01:00
Houtan Bastani b7969980d0 reporting: @page: allow for multiple titles, title_format, first args can be provided as strings or cellstr, multiple args as cellstr 2013-03-19 12:50:07 +01:00
Sébastien Villemot ea56c41b68 Merge pull request #325 from JohannesPfeifer/master
Clarification for second moments at second order and to rplot; restriction of endogenous prior to stationary models
2013-03-19 01:33:27 -07:00
Johannes Pfeifer 40e5c61ec7 Restrict endogenous prior to stationary models and allow for measurement error 2013-03-18 23:46:15 +01:00
Johannes Pfeifer 1883eb092b Clarify theoretical second moments for order=2
Eliminates warning message introduced in
4c8f3a89cc
Adds hint to approximation in table title and adds the information with
a reference to Kim/Kim/Schaumburg/Sims (2008) to manual.
closes #278
2013-03-18 22:58:06 +01:00
Johannes Pfeifer c85338f022 Add figure title for rplot 2013-03-18 22:51:54 +01:00
Houtan Bastani 0e5acda10e bug fix: fputs not a matlab function 2013-03-18 16:25:38 +01:00
Houtan Bastani aac25a88bd bug fix: substr not a matlab function 2013-03-18 16:19:48 +01:00
Houtan Bastani ec9f53a23b bug fix: single quotes 2013-03-18 16:17:29 +01:00
Sébastien Villemot cda0e03776 Add preprocessor messages (stdout, stderr) to the logfile
Closes #306
2013-03-18 13:44:04 +01:00
Sébastien Villemot ba889e853c Preprocessor interface and documentation for endogenous_prior option
Closes #321
2013-03-18 11:12:49 +01:00
Sébastien Villemot 030b7c96e0 Merge pull request #322 from JohannesPfeifer/master
Another bugfix for rplot
2013-03-18 03:10:52 -07:00
Johannes Pfeifer 1df8bf15c2 Bugfix in rplot + typo correction 2013-03-18 10:59:32 +01:00
Sébastien Villemot 4b095c2ce5 Merge pull request #319 from JohannesPfeifer/master
Bugfixes, documentation, and features
2013-03-18 01:52:43 -07:00
Johannes Pfeifer e76828bb9c Bugfix in rplot
Make other plot types take into account that variables are in rows, not
columns
2013-03-18 08:46:25 +01:00
Johannes Pfeifer 98ec0095df Add warning for variance decomposition at order=2
At order=2, the variance decomposition is still based on order=1. This
commit introduces a warning. See issue #278
(https://github.com/DynareTeam/dynare/issues/278). This information that
the variance decomposition is only triggered at periods=0 has been added
to the manual.
2013-03-17 23:36:43 +01:00
Johannes Pfeifer a090a7c35e Add endogenous priors
Add option and code for endogenous priors according to
Christiano/Trabandt/Walentin 2011, JEDC. Still needs to be integrated to
manual and pre-processor.
2013-03-17 22:51:23 +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
Johannes Pfeifer f1406487dc Make field names consistent with different analysis steps and manual
still related to #316 (https://github.com/DynareTeam/dynare/pull/316)
2013-03-17 22:47:47 +01:00
Johannes Pfeifer 4c8f3a89cc Make graph titles more expressive 2013-03-17 22:45:50 +01:00
Johannes Pfeifer 3b29e69e40 Prevent spectral density plots being triggered in identification
If options_.SpectralDensity.trigger is specified, identification would
otherwise plot graphs in every MC step
2013-03-17 15:30:49 +01:00
Sébastien Villemot c58aebd2ef Fix dyn_saveas for Octave 2013-03-16 14:21:16 +01:00
Sébastien Villemot c0029bc109 Merge pull request #316 from JohannesPfeifer/master
Change of Field name to be consistent with manual
2013-03-16 06:04:35 -07:00
Johannes Pfeifer cd26c3cebd Bugfix in dyn_saveas
Fixed a bug for the non-Octave version where only the last active figure
was printed instead of the figure with the actual handle. Octave
eps-version still needs to be taken care of.
2013-03-16 11:47:56 +01:00
Johannes Pfeifer 36ea30346c Add missing legend to plot
Add missing legend to sensitivity plot
2013-03-16 11:41:03 +01:00
Houtan Bastani 8185ee9729 reporting: @graph: accept dynDate ranges for shade and xrange and float lower/upper bound for yrange 2013-03-15 18:56:03 +01:00
Houtan Bastani b9e61f13ce dynDates: bug fix 2013-03-15 18:56:03 +01:00
Houtan Bastani 1d70daa3d9 reporting: @graph: support multiple data sets in one graph 2013-03-15 18:56:03 +01:00
Houtan Bastani d8ae22cebd reporting: cosmetic changes 2013-03-15 18:56:03 +01:00
Houtan Bastani ed5f2a8bf0 reporting: @section: check user input 2013-03-15 18:56:02 +01:00
Houtan Bastani 5715dea155 reporting: @page: check user input 2013-03-15 18:56:02 +01:00
Houtan Bastani 5aabe16a03 reporting: @report: check user input 2013-03-15 18:56:02 +01:00
Houtan Bastani e8e9b2416a reporting: @graph: check user input 2013-03-15 18:56:02 +01:00
Houtan Bastani 6b9107c1b1 reporting: @table: user input bug fix 2013-03-15 17:23:21 +01:00
Houtan Bastani 8591e7b4b7 fix typo 2013-03-15 16:43:56 +01:00
Johannes Pfeifer e18dc65871 Merge remote-tracking branch 'upstream/master' 2013-03-15 15:38:37 +01:00
Stéphane Adjemian (Ulysses) fe882dfe3d Added @dynDates/minus method. 2013-03-14 23:08:35 +01:00
Stéphane Adjemian (Ulysses) 8664a12ed1 Fixed bug. 2013-03-14 23:08:35 +01:00
Stéphane Adjemian (Ulysses) f72c45787c Cosmetic change. 2013-03-14 23:08:35 +01:00