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
84f2b50043
Block using presample option with DSGE-VAR
2015-07-08 09:21:50 +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
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
214dc74723
- Fixed bugs around analytic derivation.
...
- Fixed test routine, eliminating diffuse filter.
- Trapped incompatibility of diffuse filter with analytic derivation.
2015-04-01 09:00:51 +02:00
c7153ba2ea
Add initialization of bounds if estim_params is empty
2014-12-08 11:29:15 +01:00
6d12f2141f
Merge pull request #765 from JohannesPfeifer/analytic_derivation_message
...
Fix message in dynare_estimation_init.m
2014-11-21 22:10:30 +01:00
68b27252d7
Fix selected_variables_only option with classical forecasts
...
dyn_forecast requires that the smoother has been run on all endogenous variables, leading to crashes. Now, if this option is encountered with classical forecasts, it is overridden. Also documents this in the manual.
2014-11-16 21:08:18 +01:00
a159442935
Fix message in dynare_estimation_init.m
...
If some parameters are NaN after computing the steady state, the message erroneously said that parameters had been changed. In a check for equality, NaN is not equal to NaN. Thus, NaNs are now taken care of separately.
2014-10-29 09:02:29 +01:00
f48566aeae
Fixed prior bounds (according to the doc in master branch).
...
* Second and third positional arguments after the name of the
estimated parameter in the estimated_params block are only
considered in the optimization stage (not in the MCMC)
* Do not store bounds in bayestopt_, because bounds do not always
reflect restrictions implied by prior shapes.
* prior_bounds routine returns a structure (with fields lb and ub)
instead of a matrix.
2014-10-20 16:18:54 +02:00
3e427020d6
Cosmetic changes in headers
2014-09-21 11:51:46 +02:00
79c5e18bfe
Cosmetic changes
...
- Suppresses spurious output from dynare_estimation_1.m
- Removes unset output argument from dynare_estimation_init.m (which seems to be totally unused as it was not set at all and nobody reported any crashes)
- Corrects header of dynare_estimation_init.m
2014-09-19 16:20:43 +02:00
41d1533086
Add check that prevents datafile from having same name as mod-file
...
Otherwise, the generated m-file might be called as the datafile, leading to infinite loops. The check cannot be moved to makedataset because M_.fname is not available there.
2014-09-10 17:49:33 +02:00
6030fed4ec
Fix bug in dynare_estimation_init.m with analytic_derivation where parameter vector was not set
...
Before testing whether the steady state file changes parameter values, the parameter values were not set, leading to subsequent crashes
2014-08-14 10:47:29 +02:00
1e4c2e0881
Fixed bug affecting prior restictions (removed the extension in the argument of str2func).
2014-07-28 13:02:10 +02:00
916fe62218
Fixed bug. Test on multiple declaration of observed variables was wrong.
2014-07-24 17:53:20 +02:00
0efcef8f20
Added the possibility to declare non linear prior restrictions over estimated parameters.
...
If a file <MOD_FILE_NAME>_prior_restrictions.m exists in current folder, the value returned by this routine is
substracted to fval (likelihood-lnprior) at the end of dsge_likelihood. The arguments of this routine are: M_,
oo_, options_, dataset_ and dataset_info. Routines for writing <MOD_FILE_NAME>_prior_restrictions.m will be
provided later.
2014-07-21 12:45:49 +02:00
afb5be2067
Set options_.nobs if the new data interface is used.
2014-07-17 22:05:08 +02:00
f4f7b8d0a1
Merge pull request #660 from JohannesPfeifer/mode_file
...
Restore backward compatibility of mode_file option
2014-07-01 18:00:48 +02:00
0b82874268
Adapted madataset routine for Gsa.
2014-06-25 11:34:39 +02:00
06f74c4603
Cosmetic change.
2014-06-23 16:06:58 +02:00
800de3fcc9
Fixed bug (varobs is a field of options_).
2014-06-23 16:06:29 +02:00
efcf6bd9c0
Use dseries object in the estimation routines.
2014-06-16 17:41:59 +02:00
2caa9eb9d4
Restore backward compatibility of mode_file option
...
Displays warning and then assumes the user provided a correct mode-file. Closes #658
2014-06-01 15:08:09 +02:00
9586f1e516
Fixed bug.Variable available_extensions is a cell array.
2014-05-19 16:42:50 +02:00
211f05ac31
Added checks for the extension of the declared datafile.
2014-05-02 12:09:00 +02:00
16194cedd9
Removed calls/references to @dynDate (replaces by @dates).
2014-04-30 15:44:30 +02:00
ad0802005b
Replaced dynSeries by dseries.
2014-04-30 15:38:18 +02:00
0f62d74ee4
Fixed bug. Global options_.dataset.file must be initialized as an empty array.
2014-04-30 15:35:43 +02:00
66d08ac3bf
Merge branch 'master' into use-dynSeries
...
Conflicts:
matlab/dynare_estimation_init.m
matlab/global_initialization.m
matlab/prior_posterior_statistics.m
matlab/read_variables.m
matlab/set_prior.m
matlab/utilities/dataset/initialize_dataset.m
preprocessor/ComputingTasks.cc
2014-04-30 10:10:30 +02:00
7864e0ec82
Rewrote the error message issued when the prefilter option in used while the steady state of the observed variables is non zero (introduced in commit d82252e805).
2014-01-30 17:57:36 +01:00
d82252e805
Filter out inconsistent specification of prefiltering when observation equation is not mean zero
2014-01-30 10:07:41 +01:00
f16689e76f
Do not systematically throw an errror message if the parameters in the provided mode_file does not exactly match the set of estimated parameters.
...
If some parameters are missing in the provided mode_file and if mode_compute>0, use the prior mean as an initial condition.
2013-12-11 10:01:22 +01:00
4afcf8fdaf
Fixed typo.
2013-12-11 10:01:22 +01:00
b03697b342
Add possibility to initialize parameters from calibrated model
2013-11-06 13:50:46 +01:00
e043966906
Merge pull request #486 from JohannesPfeifer/prior_check
...
Check initial values for violation of inverse gamma prior
2013-09-30 08:23:08 -07:00
7e7cadb878
Check initial values for violation of inverse gamma prior
...
The inverse gamma distribution does not allow for the value 0, but the current check at the lower bound set LB=0 and tested for <LB instead of <=0.
Also fixes
2013-09-30 17:10:16 +02:00
1f8b4d9a86
Fix bug in display of parameter names violating the bounds condition
2013-09-30 17:02:25 +02:00
303e3845bb
Added checks about the first period (defined by the set_time command or in the datafile).
2013-09-24 16:35:49 +02:00
39408b3b05
Removed logged_data_flag and variables related to xls files.
2013-09-23 08:42:25 +02:00
c5c3b45f3c
Merge branch 'master' into use-dynSeries
2013-09-23 08:14:46 +02:00
22968497ad
Fixed bug manifesting if a model is first estimated with cova_compute=0 and if the resulting mode file is loaded after.
...
Added condition on the existence of hh in the mode_file.
2013-09-19 15:25:23 +02:00
9f63c4081f
Fixed bug affecting fs2000/fs2000_calib.mod and shock_decomp_calibrated_model/example1_calib_shock_decomp.mod.
2013-09-18 11:38:58 +02:00
a29db4175d
fixed bug with calibrated models in dyn_estimation_init.m introduced
...
in commit 952139b
2013-09-18 11:23:47 +02:00
93357d0026
Added instatiation of a dynSeries object for the dataset.
2013-09-12 17:29:40 +02:00
b680b214cc
Cosmetic changes.
2013-09-12 12:00:04 +02:00
96a46622af
Added prior mode and hyperparameters in oo_.prior structure..
2013-09-12 11:58:25 +02:00
55e3bcc5f8
Removed general transformation of the raw data, only consider the log function.
2013-09-11 11:26:35 +02:00
ddce110bf2
Merge branch 'master' into use-dynSeries
2013-09-10 22:26:12 +02:00
952139bbc9
Changes related to the treatment of the (optionally) provided mode file.
...
(1) Added more checks on the content of the provided mode file compared the the list of declared parameters (the condition on the number of parameters is not strong enough).
(2) Added a mechanism to adapt the content of the mode file if possible. For instance, if the estimated parameters are a subset of the parameters in the mode file, we only need to discard some of the parameters in the mode file.
(3) Added output argument in dynare_estimation_init, which returns the hessian matrix (hh) with the estimated mode.
2013-09-10 22:25:16 +02:00
bdde385ec6
Moved the test on the declaration of the observed variables.
2013-09-09 14:54:28 +02:00
421e7a83d5
Fixed bug.
2013-09-09 14:21:11 +02:00
1339987af5
Removed n_varobs.
2013-09-09 14:07:08 +02:00
f465949cf5
Cosmetic changes.
2013-09-09 14:05:56 +02:00
806b5a799f
Added comment.
2013-09-09 14:01:17 +02:00
e466206f6c
Removed useless block of code.
2013-09-09 13:59:06 +02:00
7756bc3af2
Removed lgyidx2varobs field (vector of integers) in options_. Added checks about the declared observed variables.
2013-09-09 13:57:34 +02:00
c824217163
Added comments.
2013-09-09 12:36:43 +02:00
907e087ea9
Consider options_.varobs as a cell of strings.
2013-09-07 16:39:04 +02:00
5abbabbd8e
Added comment.
2013-09-07 16:05:01 +02:00
46d451b34e
Moved up the section where bayestopt_.jscale is updated
2013-09-07 16:05:01 +02:00
ecd429d268
Changed the logged data section.
2013-09-07 16:05:01 +02:00
8be8bf1b1b
Recycle options_.dataset. If not empty (default) dataset field has to be a dynSeries object.
2013-09-07 16:05:01 +02:00
b4f8532bea
Merge pull request #441 from JohannesPfeifer/mode_check
...
Bugfix for mode_check and display of parameters out of bounds
2013-08-12 03:31:50 -07:00
4052d4ccaf
Cosmetic changes. Use skipline() instead of disp(' ').
2013-07-10 17:12:34 +02:00
d9579418fd
Bugfix for correct display of names of parameters outside of bounds
2013-07-06 00:20:07 +02:00
120db8f1ea
Cosmetic changes.
2013-06-24 10:24:55 +02:00
61485ab809
Fix copyright notices
2013-06-12 17:04:46 +02:00
dce8e74c33
Fix bug where initial_estimation_checks was performed on prior mean instead of specified mode
...
If the mode_file option was used, xparam1 was reloaded from the
mode-file after the initial checks, sometimes leading to very cryptic
errors. Now the initial checks already load the correct parameter
vector.
2013-04-26 21:10:36 +02:00
c121aa14b1
Remove oo_.dr.{nstatic,npred,nboth,nfwrd,nspred,nsfwrd}
...
Replace them by equivalents in M_ (and an extra one: M_.dynamic).
IMPORTANT POINT: oo_.dr.npred used to count both purely backward and mixed/both
variables. This was the cause of lots of confusion. The new M_.npred only
counts purely backward variables.
We now have the following indentities:
M_.npred + M_.nboth + M_.nfwrd + M_.nstatic = M_.endo_nbr
M_.nspred = M_.npred + M_.nboth
M_.nsfwrd = M_.nfwrd + M_.nboth
M_.ndynamic = M_.npred + M_.nboth + M_.nfwrd
2012-11-16 20:05:13 +01:00
c2c88600a1
More explicit error message when no estimated_params and no smoother
...
Closes : #222
2012-10-08 14:42:09 +02:00
cc26a1d806
Add comment for the use case when there is no estimated_params
2012-10-08 14:32:22 +02:00
d60202616a
fixed problem with penalty in estimation. Created a new global scalar:
...
objective_function_penalty_base. It is the only simple way that I
found to keep csminwel1.m to be able to handle general functions.
2012-08-28 12:17:07 +02:00
b02303cf69
Force analytic_derivation = 1;
2012-08-21 15:53:02 +02:00
b10b2160ed
Made error message if initial parameter values are outside prior bounds more explicit.
2012-07-19 12:30:41 +02:00
3414d4d468
Fix crash in estimation introduced in 1fb89a07
2012-07-06 09:50:30 +02:00
ed4d37341c
Fix problem with models where steadystate files change parameter values.
...
1) allow to compute derivatives starting from NUMERICAL derivatives of jacobian and steady state: this has a minor cost in accuracy and allow apply without errors identification and estimation with numerical derivatives;
2) added trap in dynare_estimation_init: if steadystate changes param values, automaticly shifts to numerical derivs of jacoban and steady state + analytic derivatives of all the rest;
3) bug fixes for 2nd order derivatives w.r.t. model parameters;
2012-07-05 10:14:10 +02:00
1fb89a07e9
Removed global from set_state_space.
2012-07-03 11:29:18 +02:00
3ff832ddcc
added test for too large standard deviation in beta distribution;
...
cleaned up error messages related to priors; set error trap for
initilial likelihood Nan or complex
2012-06-13 15:47:01 +02:00
a39a0b3b67
fixed bug when only variances are estimated, but no deep parameter
2012-06-13 14:08:58 +02:00
1f9cea669a
Update copyright notices
2012-06-08 18:22:34 +02:00
da9ec0f187
Estimation with analytic scores and hessian;
...
This includes re-setting the list of output arguments in objective functions
Added test function
2012-04-29 21:18:33 +02:00
e692185c6b
storing oo_.prior.mean oo_.prior.variance
...
oo_.posterior.optimization.mode oo_.posterior.optimization.variance
oo_.posterior.metropolis.mean oo_.posterior.metropolis.variance as
aggregate arrays in addition to previous storage variable by variable
2012-04-21 21:28:03 +02:00
973302e42e
fixing bugs in smoother
2012-04-18 21:15:18 +02:00
978f29f59d
Added missing field in bayestopt_ (restrict_var_list, indices for the state and observed variables).
2012-03-04 22:12:59 +01:00
f798118902
Updated test on options_.order (current version can estimate dsge models with non linear filters considering a second order approximation of the model).
2012-03-04 22:12:59 +01:00
09a3f3a084
Updates M_params only if estim_params_ has a field param_vals (not the case for a kalman_smoother)
2012-02-17 10:36:21 +01:00
ad2a3a7ecc
fixing problem in computation of initial posterior density when steady
...
state model is provided by the user
2011-12-15 17:35:27 +01:00
244e0ffb14
Corrects a minor bug related to a matrix transposition
2011-11-25 13:07:50 +01:00
e4c803d0db
fixed issues with estimation of non-stationary models. Option lik_init=2
...
is contradictory with diffuse_filter or unit_root_variables
declaration. Models with non-stationary variables, but only stationary
observed variables need diffuse_filter option and make a useless call
to kalman_filter_d (this seems better than trying to distinguish these
rare cases)
2011-11-21 12:39:02 +01:00
bcf7e88217
Do not plot the priors if option nograph is used.
2011-11-14 17:56:27 +01:00
f0efbee26d
adding short circuit operator
2011-11-02 18:34:53 +01:00
b4e7e55031
fixing bug with univariate filter
2011-11-02 17:54:48 +01:00
9ac3e7a582
fixing bug when estimating models with a steady state different from
...
zero. Introduced in commit bd00dc11d8
2011-11-01 18:23:12 +01:00
bd00dc11d8
removing 8th output argument of dynare_estimation_init and
...
corresponding seemingly useless code
2011-10-21 22:09:45 +02:00
7e3225680f
fixing logic flaw in setting lik_init as a function of diffuse filter or kalman_algo
2011-10-21 14:21:40 +02:00
ddc029394c
Clean-up last remnants of unit_root_vars ( Closes : #167 )
2011-10-14 16:22:47 +02:00
c6a9650d20
Adapt identification routines after new dynare estimation commits.
2011-09-28 20:48:47 +02:00
f1ffeb29bb
Correction of the previous commit ( 3412d5e20a). Set options_.lik_init=3 if lik_init==1 and unit_root_vars
...
keyword appears in the mod file.
2011-09-21 19:05:34 +02:00
3412d5e20a
Bug fix. Do not override option lik_init (declared in the estimation command) if unit_root_vars
...
keyword is present in the mod file, by forcing options_.lik_init to be equal to 3 (diffuse kalman
filter).
2011-09-21 18:54:11 +02:00
MichelJuillard
Johannes Pfeifer
Johannes Pfeifer
Stéphane Adjemian
Marco Ratto
Sébastien Villemot
Sébastien Villemot
Sébastien Villemot
Ferhat
Michel Juillard