f28a7a62bb
Only test non-zero measurement error covariance entries for positive definiteness
...
Otherwise, not having measurement error on one variable is not allowed during estimation
2015-12-15 22:37:08 +01:00
313ffb5f32
Merge pull request #1116 from JohannesPfeifer/Latex_cosmetics
...
Various improvements to LaTeX output
2015-12-15 17:22:00 +01:00
74283c45d2
Added unit test for priordens routine.
2015-12-15 09:14:50 +01:00
adb5a2359d
Cosmetic change.
...
Replaced 1/0 by booleans in priordens routine.
2015-12-15 08:45:13 +01:00
944ed4b42b
Fixed bug.
...
Forgot to change the name of the routine for the inverse CDF of the
Weibull distribution.
2015-12-14 23:51:50 +01:00
e53f3a67fa
Fixed unit test for prior_draw routine.
2015-12-14 23:50:56 +01:00
29a0765a48
Removed globals from prior_draw routine.
...
When initializing the persistent variables, bayestopt_ and
options_.prior_trunc need to be passed as input arguments.
2015-12-14 23:50:56 +01:00
e5894c4157
Add eps-loaders to GSA output
2015-12-14 14:09:37 +01:00
877b2cbb80
Use longtable for collinearity pattern table
2015-12-14 11:27:00 +01:00
4f9b141c76
Add eps-loaders to identification codes
2015-12-14 11:26:33 +01:00
9cea813249
Add missing packages to collect_LaTeX_Files.m and manual
2015-12-13 16:57:25 +01:00
246ea146d1
Use booktabs package for neater table output
2015-12-13 16:43:25 +01:00
aafa3e15b9
Modularize display_estimation_results_table.m
2015-12-13 16:42:52 +01:00
df4bc5c5c8
Use TeX-name for caption in shock decomposition graphs
2015-12-13 16:42:20 +01:00
482f2bdd2c
Add LaTeX output to Geweke convergence diagnostics
...
Also allows for custom header part in dyn_latex_table.m
2015-12-13 16:41:41 +01:00
837916f45b
Add LaTeX-capacities for trace_plot.m
2015-12-13 16:40:20 +01:00
88858a1fde
Cosmetic change.
...
Use logicals instead of 0/1.
2015-12-11 19:11:47 +01:00
dc1df09519
Fixed bug.
...
Wrong number of input arguments when calling gamma_specification.
2015-12-11 18:50:54 +01:00
f185d15c36
Fixed bug introduced in ed114ee8f047ee823b4dff1fe514ea4b2de180ae.
2015-12-11 18:50:54 +01:00
4670ecb9d6
Cosmetic changes.
2015-12-11 18:50:54 +01:00
d1020e6170
Fixed GSA for Weibull prior.
2015-12-11 18:50:54 +01:00
5834ed4446
Added checks for the input arguments of wblinv routine.
2015-12-11 18:50:54 +01:00
d676c67f03
Added CDF for the Weibull distribution.
2015-12-11 18:50:54 +01:00
d9cb5da148
Added unit test.
2015-12-11 18:50:54 +01:00
32475c90b0
Fixed typo in unit test.
2015-12-11 18:50:54 +01:00
1a07f95218
Fixed bug in the unit test for the mode of beta density.
2015-12-11 18:50:54 +01:00
0a4b9bfb3d
Added missing semicolon.
2015-12-11 18:50:54 +01:00
09498d30eb
Updated header doc.
2015-12-11 18:50:54 +01:00
3c42f7e362
Added try-catch blocks in unit tests.
2015-12-11 18:50:54 +01:00
8c9736f93f
Add unit tests for compute_prior_mode.m
...
(cherry picked from commit 7d9c3ee365aba25c131b423596a1446eafd9c6ec)
2015-12-11 18:50:54 +01:00
09b4928ca8
Added lower bound as an argument.
2015-12-11 18:50:54 +01:00
cfefa6951c
Changed inputs of inverse_gamma_specification routine and added unit tests.
2015-12-11 18:50:54 +01:00
887e44f2b0
Cosmetic change. Removed loop.
2015-12-11 18:50:54 +01:00
85a4a082ad
Added a routine for computing the hyperparameters of the Beta distribution from the expectation and the variance (with unit tests).
2015-12-11 18:50:54 +01:00
24848a2ba4
Added a routine for computing the hyperparameters of the Gamma density from the expectation and the variance (with unit tests).
2015-12-11 18:50:54 +01:00
02d53b5d1d
Use statistics toolbox if available for the Weibull inverse CDF.
2015-12-11 18:50:54 +01:00
a1aa2bcd6e
Changed second input of prior_bounds routine.
...
Do not pass options_ structure but only the required field (prior_trunc).
2015-12-11 18:50:54 +01:00
a859861e19
Fixed Bug.
...
Forgot to rename calls to the routine generating Weibull deviates.
2015-12-04 18:42:26 +01:00
5ef86416bf
Merge branch 'cmr-prior'
2015-12-04 15:06:28 +01:00
5fb1b1cc23
Added Weibull prior.
2015-12-04 14:50:44 +01:00
954e57c421
reporting: submodule update
2015-12-02 12:39:51 +01:00
b7e7d73177
reporting: submodule update
2015-12-01 14:34:26 +01:00
9c42ebaa56
fix bug introduced in commit 105100c7fe
2015-11-30 21:31:25 +01:00
c373d1e1be
adding new option 'fast_kalman_filter' implementing Ed Herbst 2012 approach
2015-11-28 17:38:00 +01:00
105100c7fe
remove global in dyn_forecast() and change input and output arguments
2015-11-28 17:38:00 +01:00
eb3d0cd20e
fix problem with nobs
2015-11-28 17:38:00 +01:00
09e03e7a6f
Move closing of file in mode_check.m out of the loop
...
Code will otherwise crash with more than one figure to plot.
2015-11-28 14:07:36 +01:00
2f010aa5de
Set Filter order for BK-filter in disp_moments.m to more reasonable value
2015-11-28 14:06:56 +01:00
0eba93dd71
Add missing input argument in newrat.m to prevent crashes
2015-11-28 14:06:10 +01:00
231cf19438
Fix a typo in basic_plan.m
2015-11-27 12:11:42 +01:00
83a6963286
Rewrote get_file_extension.
2015-11-26 20:53:13 +01:00
73313dc5b7
Merge pull request #1107 from FerhatMihoubi/Branch_bytecode_display_and_exogenous_in_extended_path
...
Branch bytecode display and exogenous in extended path
2015-11-19 17:09:09 +01:00
e555c8181d
Corrections in the way to handle the exogenous
2015-11-19 17:04:56 +01:00
215de7e8dc
Closes #1057 .
...
Use a static file with a list of routines to cleared before each run (because they have
persistent variables).This static file (the script
matlab/list_of_functions_to_be_cleared.m) is distributed with Dynare and
needs to be updated for each release (or each time we add/remove
persistent variables). To update the file just do:
>> clear_persistent_variables(DYNARE_MATLAB_PATH, true)
and add/commit the generated file.
2015-11-12 14:38:16 +01:00
dd2cf673d2
Updated the dates submodule.
...
The setdiff method can return a second output argument (vector of indices). If A and B are two dates objects, and if setdiff is called with two output argument:
[C, IA] = setdiff(A, B)
then we have:
C = A(IA)
2015-11-11 22:51:25 +01:00
edd275bfcc
Merge pull request #1103 from FerhatMihoubi/Branch_fix_bug_det_cond_forecast
...
Fix bugs in det_cond_forecast.m
2015-11-11 18:37:55 +01:00
6d7218f9dc
Fix bugs in det_cond_forecast.m
2015-11-11 18:35:41 +01:00
6ee212331f
submodule update: reporting
2015-11-11 15:11:31 +01:00
f3ee32690e
fixed bug introduced in commit 660fc6
2015-11-10 09:27:36 +01:00
53e9307fe2
modified too big message in evaluate_planner_objective
...
Mention the issue in the doc
2015-11-09 09:15:44 +01:00
953ac851c0
& into &&
2015-11-09 09:03:55 +01:00
660fc64471
adding check for model size before computing value of planner objective.
2015-11-09 09:03:55 +01:00
4222269324
Merge pull request #1100 from JohannesPfeifer/initial_check_par_names
...
Fix display of parameter names changed during estimation
2015-11-04 14:40:43 +01:00
cbfe1e77ce
Merge pull request #1098 from JohannesPfeifer/particle_filter_missing_check
...
Filter out datasets with missing values before particle filtering
2015-11-04 14:39:56 +01:00
1def4f357a
Fix display of parameter names changed during estimation
...
Relied on the wrong ordering
2015-10-28 15:34:14 +01:00
1f18a2468b
Deblank field names to prevent crashes during computation of posterior moments
...
Necessary after switching to indirect indexing of structure fields
2015-10-23 08:53:23 +02:00
4d573e600d
Cosmetic change.
2015-10-22 16:10:34 +02:00
aabbed0542
Fixed syntax for declaring subfields in oo_ (conditional_variance_decomposition_mc_analysis).
2015-10-22 16:10:34 +02:00
fe0f19dc37
Fixed bug. Wrong size of the state vector...
...
When dsge_likelihood enters (because of of a detected singularity) in
univariate_kalman_filter routine with correlated measurement errors.
2015-10-22 16:10:34 +02:00
af0193fe41
submodule update: reporting
2015-10-21 17:50:26 +02:00
ab893696eb
submodule update: dseries
2015-10-21 17:49:48 +02:00
fbe57c0d08
submodule update: reporting
2015-10-21 12:02:49 +02:00
554f3df659
Filter out datasets with missing values before particle filtering
2015-10-20 15:49:07 +02:00
713fc3eae8
submodule update: reporting
2015-10-19 15:14:35 +02:00
f8efe5dec0
Updated dseries submodule.
2015-10-19 14:28:20 +02:00
0b04804a73
submodule update: dseries
2015-10-19 11:10:10 +02:00
b01867c47e
Bug fix. Closes issue #1094 .
2015-10-17 17:44:25 +02:00
dc20cac3e2
Updated dseries submodule.
2015-10-17 17:43:32 +02:00
36992933d2
Merge pull request #1062 from FerhatMihoubi/cond_forecast_and_bytecode
...
Allows to compute the extended path and the conditional forecasts wit…
2015-10-16 17:12:54 +02:00
529ccd3773
submodule update: reporting
2015-10-16 15:38:20 +02:00
f5043e11b0
update reporting path
2015-10-16 15:37:39 +02:00
1d3b1ae6c3
submodule update: reporting
2015-10-15 12:33:23 +02:00
c477d25fe2
Merge pull request #1086 from JohannesPfeifer/constant_moments
...
Save posterior moments even if the distribution is degenerate
2015-10-14 22:15:09 +02:00
f12a468bb3
Merge pull request #1085 from rattoma/optimizer_number_5
...
enhancements to optimizer 5
2015-10-14 17:33:27 +02:00
f583add550
prior_posterior_function: save results to different structures, depending on which command was called
2015-10-14 11:19:12 +02:00
b4bd896b89
Merge pull request #1087 from JohannesPfeifer/resid_fix
...
Remove automatic error from resid.m when steady state does not solve.
2015-10-14 01:12:34 +02:00
61abe6dfc7
Merge pull request #1088 from rattoma/diffuse_filter_and_smoother
...
Diffuse filter and smoother
2015-10-14 01:09:08 +02:00
01329a4e0e
Merge pull request #1084 from rattoma/gsa
...
Gsa and identification commits
2015-10-13 23:59:54 +02:00
6ea5bdde34
Merge branch 'johannes_bandpass'
2015-10-13 23:56:22 +02:00
f7175745ac
Add one-sided HP filter
2015-10-13 20:50:03 +02:00
145e2d5542
preprocessor: prior_posterior_function: change option 'prior_posterior_sampling_draws' to 'sampling_draws' #1076
2015-10-13 17:41:31 +02:00
93e7be66e7
proper use on diffuse_kalman_tol in univariate diffuse filter and smoother
2015-10-13 17:28:29 +02:00
0aa7e15d58
- Once Schur stape space transformation is done, map immediately Pinf and Pstar onto the original variablesm to avoid propagation of errors related to multiple unit and zero eigenvalues.
...
- ensure that smoother and filter get the same Pstar and Pinf in diffuse steps
2015-10-13 17:26:39 +02:00
05fc096569
Make multivariate kalman filter and smoother robust to badly scaled covariance matrix of observables.
...
This avoids shifting to univariate filter in most cases.
2015-10-13 17:15:01 +02:00
c45a054c6f
modify calling structure of execute_prior_posterior_function and update doc
2015-10-13 15:13:35 +02:00
cb3804ab51
Remove automatic error from resid.m when steady state does not solve.
...
Closes #573
2015-10-13 13:45:54 +02:00
1043a019c2
Save posterior moments even if the distribution is degenerate
...
Closes #1024
2015-10-13 11:47:35 +02:00
2d59920a70
- systematically check bounds for each parameter during optimization
...
- make htol adaptive to each parameter to speed up computations
2015-10-13 11:07:01 +02:00
61d6e49dfd
Fix rank condition check, using default MATLAB tolerance (thanks to Jesper Linde and Malin Adolfson)
2015-10-13 10:45:35 +02:00
bc8f546cbb
Provisions to to show in plots the parameter value around which the sensitivity analysis with the option neighborhood_width is done.
2015-10-13 10:44:13 +02:00
Johannes Pfeifer
Stéphane Adjemian
Houtan Bastani
Michel Juillard
ferhat
Marco Ratto