stepan
/
dynare
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

adding comment explaining initialization of persistent variable 

penalty in dsge_likelihood.m and dsge_likelihood_hh.m Removed misleading
initialization code. Added call to dsge_likelihood_hh in
initial_estimation_checks to initialize persistent variable in that
function as well.
time-shift
Michel Juillard 2012-01-09 21:23:17 +01:00
parent de8962b239
commit 111347469f
3 changed files with 33 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
matlab/dsge_likelihood.m
View File

@ -127,17 +127,21 @@ function [fval,exit_flag,ys,trend_coeff,info,Model,DynareOptions,BayesInfo,Dynar
% AUTHOR(S) stephane DOT adjemian AT univ DASH lemans DOT FR
% Declaration of the penalty as a persistent variable.
persistent penalty
% Initialization of the persistent variable.
if ~nargin || isempty(penalty)
penalty = 1e8;
if ~nargin, return, end
end
if nargin==1
penalty = xparam1;
return
end
% Persistent variable 'penalty' is used to compute an endogenous penalty to
% the value 'fval' when various conditions are encountered. These conditions
% set also 'exit_flag' equal to 0 instead of 1. It is only when
% dsge_likelihood() is called by an optimizer called by
% dynare_estimation_1() that 'exit_flag' is ignored and penalized 'fval' is
% actually used.
% In that case, 'penalty' is properly initialized, at the very end of the
% present function, by a call to dsge_likelihood() made in
% initial_estimation_checks(). If a condition triggers exit_flag ==
% 0, initial_estimation_checks() triggers an error.
% In summary, an initial call to the present function, without triggering
% any condition, guarantees that 'penalty' is properly initialized when needed.
persistent penalty
% Initialization of the returned variables and others...
fval = [];

23
matlab/dsge_likelihood_hh.m
View File

@ -41,17 +41,20 @@ function [fval,llik,cost_flag,ys,trend_coeff,info] = dsge_likelihood_hh(xparam1,
% Declaration of the penalty as a persistent variable.
persistent penalty
% Initialization of the persistent variable.
if ~nargin || isempty(penalty)
penalty = 1e8;
if ~nargin, return, end
end
if nargin==1
penalty = xparam1;
return
end
% Persistent variable 'penalty' is used to compute an endogenous penalty to
% the value 'fval' when various conditions are encountered. These conditions
% set also 'exit_flag' equal to 0 instead of 1. It is only when
% dsge_likelihood_hh() is called by an newrat() called by
% dynare_estimation_1() that 'exit_flag' is ignored and penalized 'fval' is
% actually used.
% In that case, 'penalty' is properly initialized, at the very end of the
% present function, by a call to dsge_likelihood_hh() made in
% initial_estimation_checks(). If a condition triggers exit_flag ==
% 0, initial_estimation_checks() triggers an error.
% In summary, an initial call to the present function, without triggering
% any condition, guarantees that 'penalty' is properly initialized when needed.
persistent penalty
% Initialization of the returned variables and others...
fval = [];

6
matlab/initial_estimation_checks.m
View File

@ -42,6 +42,12 @@ if DynareOptions.dsge_var
[fval,cost_flag,info] = DsgeVarLikelihood(xparam1,DynareDataset,DynareOptions,Model,EstimatedParameters,BayesInfo,DynareResults);
else
[fval,cost_flag,ys,trend_coeff,info] = dsge_likelihood(xparam1,DynareDataset,DynareOptions,Model,EstimatedParameters,BayesInfo,DynareResults);
if DynareOptions.mode_compute == 5
% this call is necessary to initialized persistent variable
% 'penalty' in dsge_likelihood_hh
[fval,llik,cost_flag,ys,trend_coeff,info] = ...
dsge_likelihood_hh(xparam1,DynareDataset,DynareOptions,Model,EstimatedParameters,BayesInfo,DynareResults);
end
end
if info(1) > 0