dynare issueshttps://git.dynare.org/Dynare/dynare/-/issues2013-02-21T14:54:08Zhttps://git.dynare.org/Dynare/dynare/-/issues/274On Windows, nan and inf are incorrectly translated in <filename>.m2013-02-21T14:54:08ZSébastien VillemotOn Windows, nan and inf are incorrectly translated in <filename>.mUnder Windows 7 (and possibly Windows XP, need to check), the nan and inf values are respectively printed as 1.#IND and 1.#INF in the M-files generated by the preprocessor. This obviously leads to a crash.
Under Windows 7 (and possibly Windows XP, need to check), the nan and inf values are respectively printed as 1.#IND and 1.#INF in the M-files generated by the preprocessor. This obviously leads to a crash.
https://git.dynare.org/Dynare/dynare/-/issues/273improve the implementation of linsolve for Octave2013-02-21T14:54:08ZSébastien Villemotimprove the implementation of linsolve for OctaveCommit d32e076b77e4e8ad01d88b94b71a7c1a5763d7cd introduced a minimal implementation of linsolve for Octave. We should probably provide a better one (maybe through an Oct-file).
Commit d32e076b77e4e8ad01d88b94b71a7c1a5763d7cd introduced a minimal implementation of linsolve for Octave. We should probably provide a better one (maybe through an Oct-file).
https://git.dynare.org/Dynare/dynare/-/issues/272graph_format with multiple formats2013-02-21T14:55:30ZSébastien Villemotgraph_format with multiple formatsCurrently the option graph_format only allows only one format at a time to be selected.
This can be a strong limitation (assume one wants both eps for latex and fig to handle and edit plots).
We should allow multiple entries with a prep...Currently the option graph_format only allows only one format at a time to be selected.
This can be a strong limitation (assume one wants both eps for latex and fig to handle and edit plots).
We should allow multiple entries with a preprocessor syntax
graph_format =(eps,fig)
The preprocessor would define the variable in options_ as:
options_.graph_format = char('eps', 'fig' );
which would properly be handled in dyn_saveas by changing strcmp with strmatch.
https://git.dynare.org/Dynare/dynare/-/issues/271correct initialization of simulation variables2014-04-08T15:53:59ZSébastien Villemotcorrect initialization of simulation variables1) the current implementation process on the fly the initialization information provided by initval, endval and shocks, creating along the way oo_.endo_simul, oo_.exo_simul and oo_.exo_det_simul (information provided by histval is now co...1) the current implementation process on the fly the initialization information provided by initval, endval and shocks, creating along the way oo_.endo_simul, oo_.exo_simul and oo_.exo_det_simul (information provided by histval is now correctly stored in M_ before processing)
2) this implementation is prone to errors and forces users to respect a strict order of declaration. Simple checks of compliance with this order prevents users some legitimate usage such as macro loops
3) the solution is to record the initialization information in fields of M_ and to create/initialized oo_.endo_simul, oo_.exo_simul and oo_.exo_det_simul just before they are used.
4) set_shocks, make_y and make_ex should be rewritten and, for the last two, their name changed.
https://git.dynare.org/Dynare/dynare/-/issues/270Determinant computation in marginal_density.m looks sub-optimal2015-08-06T13:33:30ZSébastien VillemotDeterminant computation in marginal_density.m looks sub-optimalIn marginal_density.m, the determinant is computed (at three places) using the det() function. This function is known to be less precise than a Cholesky-based determinant computation when the determinant is close to zero.
Stéphane think...In marginal_density.m, the determinant is computed (at three places) using the det() function. This function is known to be less precise than a Cholesky-based determinant computation when the determinant is close to zero.
Stéphane thinks that the Cholesky was used before, so we need to understand why this has changed, and therefore whether we can go back to the Cholesky.
Issue reported by Gilles Bélanger (Ministère des Finances du Québec)
https://git.dynare.org/Dynare/dynare/-/issues/269add dynare_version() to the manual2013-02-21T14:55:30ZSébastien Villemotadd dynare_version() to the manualhttps://git.dynare.org/Dynare/dynare/-/issues/268In Dynare++, use libmatio instead of home-made routines for MAT files2013-02-21T14:55:30ZSébastien VillemotIn Dynare++, use libmatio instead of home-made routines for MAT fileshttps://git.dynare.org/Dynare/dynare/-/issues/267design an interface for assigning moment/irf contraints (e.g. sign constraints)2016-05-17T11:03:52ZSébastien Villemotdesign an interface for assigning moment/irf contraints (e.g. sign constraints)Currently there is no way of setting a set of constraints to irfs or moments of DSGE model to feed to some calibration procedure like Monte Carlo filtering in the sensitivity analysis toolbox.
Indeed the threshold_redform option in sens...Currently there is no way of setting a set of constraints to irfs or moments of DSGE model to feed to some calibration procedure like Monte Carlo filtering in the sensitivity analysis toolbox.
Indeed the threshold_redform option in sensitivity analysis would only allow to set, e.g., the SAME sign constraint to a set of irfs, which is not very handy.
So, we may think to the following syntax for sign constraints in moments:
moment_calibration;
y, y(-1), +; [i.e. auto-correlation with 1 lag]
c, g(-3), +; [i.e. auto-correlation with generic lags]
i, g, -; [i.e. cross-correlation, contemporaneous]
end;
for irfs [y c i are endogenous, em eg are exogenous shocks]:
irf_calibration;
y, em, -;
c, eg, +;
i, eg, -;
end;
or the following for more detailed numeric boundaries, in principle mixed with simple sign constraints
moment_calibration;
y, y(-1), [0.5 1]; [auto-correlation with 1 lag]
c, g(-3), +; [auto-correlation with generic lags]
i, g, [-1, -0.5]; [cross-correlation, contemporaneous]
end;
irf_calibration;
y, em, [-0.01 0];
c, eg, [0 0.02];
i, eg, -;
end;
The sign constraint can also be defined using
[0 inf] for +
[-inf 0] for -
JRC-2014M3https://git.dynare.org/Dynare/dynare/-/issues/266allow inf in threshold redform option of sensitivity analysis2013-02-21T14:55:30ZSébastien Villemotallow inf in threshold redform option of sensitivity analysisCurrently, only real numbers are accepted by the option
threshold_redform of the sensitivity toolbox.
However, it would be handful to allow inf as input, e.g. if one wants to filter positive numbers in an irf, one would like to use a syn...Currently, only real numbers are accepted by the option
threshold_redform of the sensitivity toolbox.
However, it would be handful to allow inf as input, e.g. if one wants to filter positive numbers in an irf, one would like to use a syntax [0 inf].
https://git.dynare.org/Dynare/dynare/-/issues/265add @#ifndef to macroprocessor2013-02-21T14:55:30ZSébastien Villemotadd @#ifndef to macroprocessorfor gpm
for gpm
https://git.dynare.org/Dynare/dynare/-/issues/264add checksum to download section of web site2013-02-21T14:55:30ZSébastien Villemotadd checksum to download section of web sitehttps://git.dynare.org/Dynare/dynare/-/issues/263latex variable with exponent causes compilation error2020-01-13T17:40:52ZSébastien Villemotlatex variable with exponent causes compilation errorSee: http://www.dynare.org/phpBB3/viewtopic.php?f=1&t=3846
Quick fix is to require users to place brackets around latex
variables containing exponents (and probably underscores)
Complete fix is to place braces around variables with le...See: http://www.dynare.org/phpBB3/viewtopic.php?f=1&t=3846
Quick fix is to require users to place brackets around latex
variables containing exponents (and probably underscores)
Complete fix is to place braces around variables with lead/lags and those variables with no lead/lag but that are followed by an exponent. In the user's case, we would output: B^problem_t-1^\rho
4.6Houtan BastaniHoutan Bastanihttps://git.dynare.org/Dynare/dynare/-/issues/262unify names for maxit options2019-02-08T08:31:00ZSébastien Villemotunify names for maxit optionsSeveral functions need an optional maximum number of iterations. There is currently no unified name style for these options.
Until now, we tried to have exactly the same name for the option in the language and the options_ fieldname.
...Several functions need an optional maximum number of iterations. There is currently no unified name style for these options.
Until now, we tried to have exactly the same name for the option in the language and the options_ fieldname.
In that case, I wonder if we should not have always maxit in the language and specialized the fieldname with a reference to the function. For example
maxit in steady -> options_.solve_maxit
Stéphane Adjemianstepan@adjemian.euStéphane Adjemianstepan@adjemian.euhttps://git.dynare.org/Dynare/dynare/-/issues/261Option to require that all endo/exo are initialized in initval2016-06-02T10:58:44ZSébastien VillemotOption to require that all endo/exo are initialized in initvalImplement a new option to initval command, "all_values_required". The preprocessor would stop and reports which endogenous or exogenous variable has not been explicitly initialized if any (instead of using a default zero value).
Request...Implement a new option to initval command, "all_values_required". The preprocessor would stop and reports which endogenous or exogenous variable has not been explicitly initialized if any (instead of using a default zero value).
Request from the IMF for GIMF
4.5Houtan BastaniHoutan Bastanihttps://git.dynare.org/Dynare/dynare/-/issues/260implement trust-region method for non-linear solver2014-02-04T16:56:31ZSébastien Villemotimplement trust-region method for non-linear solverThe code in solve1.m to deal with the case where the Jacobian is near singular is unsatisfactory. We need to implement a trust region method as discussed in Nocedal and Wright (2006. This could also be done for the non-linear solver of b...The code in solve1.m to deal with the case where the Jacobian is near singular is unsatisfactory. We need to implement a trust region method as discussed in Nocedal and Wright (2006. This could also be done for the non-linear solver of bytecode
4.5Sébastien VillemotSébastien Villemothttps://git.dynare.org/Dynare/dynare/-/issues/259Shock decomposition on calibrated models fails2013-06-03T14:47:59ZSébastien VillemotShock decomposition on calibrated models failsThe bayestopt_ structure is uninitialized.
The bayestopt_ structure is uninitialized.
4.4MichelJuillardMichelJuillardhttps://git.dynare.org/Dynare/dynare/-/issues/258Make flexible the location of user configuration file with global options2013-02-21T14:56:04ZSébastien VillemotMake flexible the location of user configuration file with global optionsCommit 41def7caaaabc70b659ea528a315122ec53366ae introduced the possibility to have a user config file executed after global_initialization.m. The location of this file is currently hard-coded.
The user should be able to choose the locat...Commit 41def7caaaabc70b659ea528a315122ec53366ae introduced the possibility to have a user config file executed after global_initialization.m. The location of this file is currently hard-coded.
The user should be able to choose the location of this file. The interface would be a new stanza/field in the already-existing ~/.dynare config file.
Other such "hook" files could be added at other places in the future.
4.3https://git.dynare.org/Dynare/dynare/-/issues/257Broken estimation of models with non-diagonal covariance matrix for shocks an...2013-03-21T11:56:48ZSébastien VillemotBroken estimation of models with non-diagonal covariance matrix for shocks and calibrated correlationSee this thread:
http://www.dynare.org/pipermail/dev/2012-May/002015.html
See this thread:
http://www.dynare.org/pipermail/dev/2012-May/002015.html
4.3https://git.dynare.org/Dynare/dynare/-/issues/256allow for non-parallel related configuration items in conffile2013-02-21T14:56:04ZSébastien Villemotallow for non-parallel related configuration items in conffileModify preprocessor/ConfigFile.cc to accept non-parallel-related configuration commands. Add a new section, e.g. [hooks] with, for the moment only one, option: global_initialization. i.e., something like:
[hooks]
global_initialization_f...Modify preprocessor/ConfigFile.cc to accept non-parallel-related configuration commands. Add a new section, e.g. [hooks] with, for the moment only one, option: global_initialization. i.e., something like:
[hooks]
global_initialization_file = path/to/file.m
Write /path/to/file.m to var in modfile.m then execute it at the end of global_initialization.m if the var is not empty.
Then, revert change 41def7caaaabc70b659ea528a315122ec53366ae
4.3https://git.dynare.org/Dynare/dynare/-/issues/255Check for consistent use of dataset_.descriptive.mean vs. bayestopt_.mean_var...2016-03-23T13:17:57ZSébastien VillemotCheck for consistent use of dataset_.descriptive.mean vs. bayestopt_.mean_varobs in smoother and estimationCurrently, both structures are still present. prior_posterior_statistics.m sets
bayestopt_.mean_varobs = dataset.descriptive.mean';
to have both. Thus, use should be made consistent and storage be confined to one structure. As the mea...Currently, both structures are still present. prior_posterior_statistics.m sets
bayestopt_.mean_varobs = dataset.descriptive.mean';
to have both. Thus, use should be made consistent and storage be confined to one structure. As the mean of the data belongs to the dataset and not the estimation, the dataset_ structure is better suited than bayestopt_.