38e1ec6b1e
Fixed bug.
...
Routine forcst was crashing if at least one of the observed variables,
with measurement error, was not in varlist.
2016-08-23 17:22:41 +02:00
5df056b38f
Fixed bug (conf_sig).
...
Use options_.forecasts.conf_sig instead of options_.conf_sig.
2016-08-23 17:22:41 +02:00
dd22b9609e
Add function generate_trace_plots.m that plots trace plots for all estimated parameters
2016-08-23 17:22:41 +02:00
1befd3b515
Make trace plots uniquely identified by adding chain number to save names
2016-08-23 17:22:41 +02:00
08d53206a9
Port textwidth figure format change to dynare_estimation_1.m
2016-08-23 17:22:41 +02:00
8341ac33f8
Document treatment of measurement error in forecasts
...
Also removes wrong info on HPDTotalsup/inf
2016-08-23 17:22:41 +02:00
894090e6be
Add missing 1 to min operator in TeX_loader
2016-08-23 17:22:41 +02:00
68e32e8e25
map_calibration.m: Remove manual setting of xticks
...
Clutters x-axis too much
2016-08-23 17:22:41 +02:00
bf725787e6
Only generate forecast with measurement error if observed variables are requested
2016-08-23 17:22:41 +02:00
eaaa0e5045
Fix TeX-names in GSA when mapping is over all parameters, not just structural ones
2016-08-23 17:22:41 +02:00
0d7d63fe9b
Fix duplicate TeX-label for IRFs
2016-08-23 17:22:41 +02:00
054c9bde76
Fix variable naming in simultxdet.m
2016-08-23 17:22:41 +02:00
0f72e6ed63
Account for measurement error in non-MCMC forecasts
2016-08-23 17:22:41 +02:00
46450c6f1f
Change textwidth depending on number of subplots
...
Prevents overblown figures
2016-08-23 17:22:41 +02:00
7050cab9c9
Consider measurement error for uncertainty bands in Bayesian estimation
2016-08-23 17:22:41 +02:00
daa9d15c75
Improve readability of plots generated by pm3_core.m
...
- Do not plot zero axis if plots are not centered around it
- Limit decimal points of yticks
2016-08-23 17:22:41 +02:00
84ab37fd7a
Remove globals from forcst.m
2016-08-23 17:22:41 +02:00
424050f380
Make sure IRFs only have full periods as xticks
2016-08-23 17:22:41 +02:00
73afe7d02f
Change fixed LaTeX figure width into \textwidth scaling
...
Improves readability
2016-08-23 17:22:41 +02:00
9a3414f652
Add missing deblank to conditional_variance_decomposition_mc_analysis.m which crashes otherwise if variables have different name length
2016-08-23 17:22:41 +02:00
de383a263e
Prevent TeX_binder from including itself
2016-08-23 17:22:41 +02:00
425ada90bb
Fix traceplot command in TeX unit test
2016-08-23 17:22:41 +02:00
ab35020e97
Merge pull request #1224 from JohannesPfeifer/ms_doc
...
Document options of MS code after interface change
2016-08-23 16:57:06 +02:00
9653568fb6
Merge pull request #1247 from JohannesPfeifer/dyn++timing
...
Clarify the timing convention in dynare++
2016-08-23 16:45:26 +02:00
5b60ffa0f7
Option fast_kalman_filter is also available with kalman_algo=0.
2016-08-22 23:31:02 +02:00
73d0a82267
Add comment on logic of singularity testing in univariate_kalman_filter_d.m
...
See discussion on mailing list 18/06/2016
2016-08-22 19:24:35 +02:00
8dc96cafa4
If F is identically 0 in Kalman filter, discard parameter draw instead of treating current observation as unobserved
...
See discussion on mailing list 18/06/2016
2016-08-22 19:24:35 +02:00
fa3a71bcae
Make warning message in dynare_estimation_1.m more informative
2016-08-22 19:24:35 +02:00
75e52ed71d
Make sure fast_kalman_filter is only used with diffuse_periods=0
...
Closes #1234
2016-08-22 19:24:35 +02:00
af72c1bd22
Add missing comment sign in DsgeSmoother.m
2016-08-22 19:24:35 +02:00
960aeb0d8d
Add fast Kalman filter to stationary unit tests
2016-08-22 19:24:35 +02:00
20e5d76d54
Fix missing negation in check for fast Kalman filter
2016-08-22 19:24:35 +02:00
a7e3ecea63
Add diffuse Kalman filter unit tests
2016-08-22 19:24:35 +02:00
06ff0c7bb6
Factorize Kalman filter unit tests
2016-08-22 19:24:35 +02:00
36ccec75ce
Add missing constant 2*pi to missing_observations_kalman_filter_d.m
2016-08-22 19:24:35 +02:00
63b8c4d2dd
Fix smoother for univariate filter with (correlated) measurement error
2016-08-22 19:24:35 +02:00
f61453211c
Correct header of missing_DiffuseKalmanSmootherH3_Z.m and make sure H is vector
2016-08-22 19:24:35 +02:00
178bf4165c
Make sure Zflag for augmented state space is correctly set when measurement error is correlated
2016-08-22 19:24:35 +02:00
f95619b46a
Add header info to univariate_kalman_filter.m
2016-08-22 19:24:35 +02:00
69c5575e19
Add smoother checks to Kalman unit tests
2016-08-22 19:24:35 +02:00
7c96e22985
Fix bug in dsge_likelihood.m where diagonal of measurement error covariance matrix was not correctly read out
...
This is a prerequisite for the univariate filter that operates along a vector for H instead of along the diagonal. Because of this, the likelihood in the presence of measurement error was wrong
2016-08-22 19:24:35 +02:00
6368636ee0
Move Kalman filter options consistency checks to dynare_estimation_init.m
2016-08-22 19:24:35 +02:00
8aaca040b3
Expand check of fast_kalman_filter for presence of block-option
2016-08-22 19:24:35 +02:00
bffea9b426
Fix error code for inf-likelihood
2016-08-22 19:24:35 +02:00
915b437efc
Add correctness check for Kalman filter unit tests
2016-08-22 19:24:35 +02:00
077b725a52
Correct comment in _dynamic files that misstated columns of derivative matrices
2016-08-22 19:24:35 +02:00
e939d51d58
display_problematic_vars_Jacobian.m: Fix bug where exogenous variables in dynamic Jacobian were not accounted for Prevents crashes in debug mode
2016-08-22 19:24:35 +02:00
b201ca4e4b
Add proper measurement error to Kalman filter test files
2016-08-22 19:24:35 +02:00
ed22b6c4f8
Make datatomfile.m datafiles save name of creator and date
2016-08-22 19:24:35 +02:00
15f95cec4a
Add comments to Kalman filtering routines
2016-08-22 19:24:35 +02:00
Stéphane Adjemian (Hermes)
Johannes Pfeifer
Houtan Bastani