15b47ef50b
Account for cases where estim_params_ is empty or has only been filled with field full_calibration_detected
2015-07-21 11:41:39 +02:00
84e04522bf
Account for options_.nk potentially being empty.
2015-07-21 10:41:47 +02:00
9b8077fc28
Merge branch 'master' into new_ep
2015-07-21 09:30:58 +02:00
80253835a7
Merge pull request #738 from JohannesPfeifer/Kalman
...
Kalman
2015-07-20 15:02:40 +02:00
f94910173d
Merge pull request #937 from JohannesPfeifer/graph_fix
...
Various fixes related to graphs and eps-TeX-loaders
2015-07-20 14:36:20 +02:00
2cb58be875
Merge pull request #969 from JohannesPfeifer/documentation
...
Improve documentation of nobs, first_obs, presample, and mode_check
2015-07-20 14:27:40 +02:00
c15e8f65c7
Merge pull request #769 from JohannesPfeifer/filtered_vars_ML
...
Do not create classical filtered variables at posterior mean if Bayesian...
2015-07-20 14:21:30 +02:00
2269b38465
Merge pull request #740 from JohannesPfeifer/relative_irf
...
Fix and document relative_irf
2015-07-20 14:20:00 +02:00
8045217773
Merge pull request #729 from JohannesPfeifer/complex_jacobian
...
Deal with debugging info on complex Jacobians
2015-07-20 14:08:33 +02:00
e174f5f00a
Merge pull request #726 from JohannesPfeifer/steady_linear
...
Prevent incorrect steady state from being accepted if model is not truly linear
2015-07-20 14:06:03 +02:00
4b3550efbc
Merge pull request #691 from JohannesPfeifer/var_decomp_unit
...
Fix dsge_simulated_theoretical_variance_decomposition.m for non-stationa...
2015-07-20 14:03:22 +02:00
9634ff2098
Merge pull request #698 from JohannesPfeifer/Fix_steady_state_estimation
...
Fix bug in dynare_estimation_init.m if steady state does not solve
2015-07-20 13:57:37 +02:00
86322bcb3a
Fix typo in dsge_likelihood.m
...
blkdiag command is not written with ck
2015-07-20 13:10:31 +02:00
b0bbab68f3
Add likelihood value to mode-file
2015-07-20 13:06:22 +02:00
d6bc85fdc6
Improve error message for incompatible filter/Kalman option
2015-07-20 13:06:22 +02:00
bb0446873f
Add initial likelihood value to debugging info
2015-07-20 13:06:21 +02:00
ce3dc068c3
Merge pull request #692 from JohannesPfeifer/imcforecast
...
Account for initial values when making conditional forecasts
2015-07-20 10:18:43 +02:00
2c8ee55d16
Revert "adding provision for histval in conditional_forecasts"
...
This reverts commit cb9a0ef064
2015-07-20 10:16:03 +02:00
cb9a0ef064
adding provision for histval in conditional_forecasts
2015-07-19 17:34:06 +02:00
edbd1ef374
fixing typo
2015-07-13 11:51:07 +02:00
dd8bd78562
Updated dseries submodule.
2015-07-09 16:03:47 +02:00
29589c062f
Moved mode_check data file (see commit fe41b64c84) in a subfolder.
2015-07-08 17:18:27 +02:00
fe41b64c84
Save on disk the data produced by mode_check.
2015-07-08 16:05:38 +02:00
69b7d8024c
submodule update: reporting
2015-07-08 11:43:19 +02:00
84f2b50043
Block using presample option with DSGE-VAR
2015-07-08 09:21:50 +02:00
d78ed888e5
Added a function to evaluate the accuracy of the linear approximation
...
when solving perfect foresight models.
2015-07-07 17:55:41 +02:00
ee44dd5fea
Updated header.
2015-07-07 17:55:41 +02:00
316f394e06
Removed unused variables from sim1.m.
2015-07-07 17:55:41 +02:00
17f3583151
Allow linear_approximation option with stack_solve_algo=7.
2015-07-07 17:55:41 +02:00
27922a349c
Fixed typo.
...
The value of jacobian_flag was not passed to fsolve.
2015-07-07 17:55:41 +02:00
8007f508a2
New option linear_approximation for perfect foresight models.
...
This approach only requires one evaluation of the dynamic model (and its
jacobian) instead of T (the number of perdiods). Also (because the model
is linear) the equilibrium paths are obtained by inverting the jacobian
of the stacked equations (no need for a Newton algorithm).
Only available with stack_solve_algo==0 (which is the default algorithm
for solving perfect foresight models).
If possible, the option is triggered automatically if the model is
declared linear.
TODO:
* Write a linear version of perfect_foresight_problem routine.
* Evaluate the approxilation error (just need to evaluate the system of
stacked non linear equations).
2015-07-07 17:55:41 +02:00
3640b2b511
minor change to perfect_foresight_solver_core.m
2015-07-07 16:42:30 +02:00
ca6da942f2
Merge pull request #968 from JohannesPfeifer/Typos
...
Fix typos in error message in stochastic_solvers.m
2015-07-06 09:37:21 +02:00
90a6bb991c
Fix typos in error message in stochastic_solvers.m
2015-07-06 08:53:54 +02:00
3d4d3ff2cb
Replace call to cholcov by call to chol
...
Closes #964
2015-07-06 08:21:09 +02:00
3b142de609
submodule update: reporting
2015-07-02 11:16:10 +02:00
73781eafcf
submodule update: reporting
2015-07-02 10:15:57 +02:00
8ab0a60164
submodule update: reporting
2015-07-02 10:08:34 +02:00
cccb0995ff
submodule update: reporting
2015-07-01 15:07:31 +02:00
a51e7251e1
submodule update reporting
2015-06-26 15:42:06 +02:00
23ff65d15a
Updated dynare_config routine (initialization of the dseries and dates classes).
2015-06-24 19:02:42 +02:00
6d2f049637
Moved routines and updated submodules dseries and dates.
2015-06-24 18:54:51 +02:00
71700dff76
Merge pull request #942 from JohannesPfeifer/TaRB_integration
...
Integrate the TaRB-algorithm into Dynare
2015-06-19 14:30:56 +02:00
3fd010ff91
Merge pull request #957 from JohannesPfeifer/fs2000_steady_state
...
Update unit tests to use correct steady state files
2015-06-18 14:25:01 +02:00
49dbd4e091
Merge pull request #960 from JohannesPfeifer/calib_smoother
...
Filter out case where estimated_params block is present but smoother on calibration is requested
2015-06-18 12:15:19 +02:00
1cd2df568a
Merge pull request #956 from JohannesPfeifer/dyn_mex_windows
...
Fix bugs in dyn_mex.m under Windows
2015-06-18 11:48:12 +02:00
73e1ddf7f0
Merge pull request #961 from rattoma/gsa
...
Fix bug for models with leads and legs in exo variables
2015-06-18 11:32:37 +02:00
bf707c1975
Merge pull request #962 from JohannesPfeifer/bvar_irf_fix
...
Port bugfix #920 for Bayesian IRFs to BVAR routine
2015-06-18 11:27:55 +02:00
770250aee9
submodule update: reporting
2015-06-17 12:58:03 +02:00
ecd03815c8
Port bugfix #920 for Bayesian IRFs to BVAR routine
...
Fixes plotting of IRFs where IRFs turn from negative to positive
2015-06-15 19:31:09 +02:00
44fd0b37f9
reporting: submodule update
2015-06-11 11:55:31 +02:00
a39c26060e
Fix bug for models with leads and legs in exo variables
2015-06-10 15:57:30 +02:00
864b5f17bc
submodule update: dates
2015-06-10 09:28:07 +02:00
22b1aa7a0d
Filter out case where estimated_params block is present but smoother on calibration is requested
...
In this case, xparam1 will be empty while estim_params_ does not have conformable dimensions.
2015-06-09 14:51:32 +02:00
ac7b8a691e
Fix evaluation of residuals stored in info
...
They were wrongly based on the initial guess
2015-06-08 18:10:58 +02:00
a5e60834ca
Ascertain that only correct steady states for linear models are accepted
...
Without the check, incorrect steady states were accepted. This happened for example if the model was not linear in the first place and solving a wrongly presumed linear equation system did not work.
2015-06-08 18:09:39 +02:00
52f9d164ac
Add routine for creating parameter table
2015-06-08 17:59:57 +02:00
f1d81dbf90
Account for latex_original file in collect_LaTeX_Files.m
2015-06-08 17:59:57 +02:00
45bbd460e7
Add routine collect_LaTeX_Files.m that integrates non-standalone TeX-codes into one file that can be compiled
2015-06-08 17:57:16 +02:00
bcc1008d3d
Only create eps-loaders if eps-figures have been created (leaves table TeX-output unaffected)
2015-06-08 17:57:15 +02:00
be567c92a3
Fix description of generating function in eps-loader in dynare_estimation_1.m
2015-06-08 17:57:15 +02:00
200187a2c9
Fix TeX-loader of PosteriorIRF.m
...
It referred to the wrong files and always loaded only one figure, even if several were created.
2015-06-08 17:57:14 +02:00
66a840b4d5
Add TeX-loader to forecast_graphs.m used with dyn_forecast.m
2015-06-08 17:57:14 +02:00
3a18b72f1b
Add TeX eps-loader to graph_decomp.m called in shock_decomposition.m
2015-06-08 17:57:14 +02:00
efb23bbca0
Add captions to pm3.m TeX-output
...
Otherwise, it is impossible to know that is displayed
2015-06-08 17:57:13 +02:00
26f6f7eb3b
Fix bug in forecast_graphs.m when var_list_ was empty
...
In this case, var_list was not set to M_.endo_names because of a typo
2015-06-08 17:57:13 +02:00
6f7265337b
Fix saving of graphs in forecast_graphs.m
...
Did not use the M_.dname folder, but rather fname
2015-06-08 17:57:12 +02:00
a339164b77
Eliminate global variables from dyn_forecast.m and add header
2015-06-08 17:57:12 +02:00
8385e813ce
Fix display of ML forecasts with Matlab 2014b graphics interface
...
Explicit colours need to be specified
2015-06-08 17:57:11 +02:00
c2f07c720f
Add check whether tarb_new_block_probability is between 0 and 1
2015-06-08 16:49:39 +02:00
3bf13dd53b
Add functionality for TaRB
2015-06-08 16:44:44 +02:00
0769fd1cce
Add verbosity and SaveFile options to some optimizers
...
Also moves additional optimizer-related files to corresponding subfolder
Closes #894
2015-06-08 16:44:44 +02:00
ccbb5fa7fb
Fix relative_irf option for higher order
...
Ex-post scaling is only allowed if linearity holds
2015-06-08 16:39:07 +02:00
3c9b031bd1
Fix bug in dynare_estimation_init.m if steady state does not solve
...
There was no check in dynare_estimation_init.m whether the steady state solved, but the noconstant option was set nevertheless. This gave rise to cases where a steady state file returned 0 for the observables, but the steady state did not exits. Due to later using use_calibration, this problem with the initial values was not detected, but the observation equation featured no constant although it should. The solution is to move the use_calibration option before the first steady state computation and then issue an error if the parameterization does not work
2015-06-08 16:35:50 +02:00
1c01bacf36
Fix bug in dyn_forecast.m where forecasts after an initval block where computed at the initial values and not at the steady state
...
This contradicted the manual (histval must be used to do this). Also clarifies the manual on this.
2015-06-08 15:48:46 +02:00
58387f00e9
Make conditional forecasts compatible with histval
2015-06-08 15:42:14 +02:00
ad66a7bb35
Account for initial values when making conditional forecasts
...
- Closes #655
- Implements unit test
- Saves distribution of controlled exogenous variables
- Adds information that the full path of constrained variables must be specified
2015-06-08 15:42:14 +02:00
5f40285307
Take care of more requested plot periods than unconditional forecast periods
...
Also takes care of initial period that was neglected
2015-06-08 15:33:54 +02:00
5d8da65e6d
Add check for unique data entries in dataset
...
Leads to cryptic crashes if not filtered out
2015-06-08 11:16:40 +02:00
41155369ac
Merge pull request #955 from JohannesPfeifer/identification_fixes
...
Various fixes to GSA and Identification
2015-06-08 10:17:24 +02:00
249c2ca7ad
Force sampling from uniform distribution for ML in stab_map_.m
2015-06-08 09:45:09 +02:00
c864d9facb
Fix histvalf.m for steady state files
...
In this case, ex0_ is required, but it was not set global, leading to crashes
2015-06-07 17:28:16 +02:00
13134e0c80
Make check_list_of_variables.m reflect that varobs is now a cell array.
2015-06-07 17:25:04 +02:00
c39331d526
Fix bugs in dyn_mex.m under Windows
...
Fixes string comparison and copy and paste error.
2015-06-07 14:07:11 +02:00
e07607011c
Merge pull request #710 from JohannesPfeifer/estimation_init_analytic_derivation
...
Fix bug in dynare_estimation_init.m with analytic_derivation where param...
2015-06-06 23:17:10 +02:00
79ca991b02
Add initialization of c1 to sylvester3.m
...
Prevents crashes in older Matlab versions for pathological cases (e.g. if p=0)
2015-06-06 18:45:05 +02:00
27622a29da
Fix initialization of parameter bounds for ML case in stab_map_.m
2015-06-06 17:40:37 +02:00
68c9c3e9ab
Account for ML in map_ident_.m
...
In this case, pdraws is empty. Closes #676
2015-06-05 20:49:43 +02:00
4e6ce64081
Account for possibility of ML estimation in stab_map_.m
2015-06-05 20:48:43 +02:00
a269bed26e
Account for fact that varobs is row cell now, not char column array in GSA
2015-06-05 20:47:51 +02:00
88ae542a8a
Fix copy and paste error in set_shocks_param.m introduced in 3ea44d47cf
2015-06-05 20:46:46 +02:00
1232d51765
Cosmetic fixes to stab_map_.m
...
Also fixes initialization of variables; closes #764
2015-06-05 19:22:03 +02:00
446b1f6dc1
Fix bug in identification when estimating only standard deviations
...
Closes #945
2015-06-05 18:22:45 +02:00
3ea44d47cf
Port changes from #511 to gsa/set_shocks_param.m
...
Closes #662
2015-06-05 16:56:32 +02:00
01cb238a22
Account for NaN in graphs during identification if ML is used
2015-06-05 16:40:01 +02:00
d09bf195c9
Make sure diffuse_filter option is correctly passed in dynare_identification.m
...
Closes #677
2015-06-05 16:39:17 +02:00
0b99367648
Fixed bug in makedataset (do not use 'exist' function, cannot make the difference between xls and xlsx files).
2015-06-05 16:35:05 +02:00
11c84b733f
Fixed bug.
...
isoctave is not in the path when calling add_path_to_mex_files.
2015-06-05 11:32:44 +02:00
6196501f37
Updated dseries submodule (fix for octave and big csv files loading).
2015-06-04 18:38:37 +02:00
c782ca3686
adding occbin original files from 20140630
2015-06-04 14:18:46 +02:00
Johannes Pfeifer
Michel Juillard
Johannes Pfeifer
Stéphane Adjemian (Charybdis)
Houtan Bastani
Marco Ratto