Dynare forums

Hi,
is there someone who can help me how to fix this bug?
I've been estimating my model and it works. But when I add the command «moments_varendo» to get endogenous variables moments, it gives only the posterior distribution of parameters and shock; Then blocks with the message :

Warning: Divide by zero. This warning will be removed in a future release.
Consider using DBSTOP IF NANINF when debugging.
> In dsge_simulated_theoretical_conditional_variance_decomposition at 71
In posterior_analysis>job at 75
In posterior_analysis at 34
In compute_moments_varendo at 110
In dynare_estimation_1 at 1081
In dynare_estimation at 62
In model1 at 120
In dynare at 132
??? Reference to non-existent field 'sigma_e_is_diagonal'.

Error in ==> dsge_simulated_theoretical_conditional_variance_decomposition at 120
StateSpaceModel.sigma_e_is_diagonal = M_.sigma_e_is_diagonal;

Error in ==> posterior_analysis>job at 75
[nvar,vartan,NumberOfFiles] = ...

Error in ==> posterior_analysis at 34
oo_ = job(type,SampleSize,arg1,arg2,arg3,options_,M_,oo_);

Error in ==> compute_moments_varendo at 110
oo_ = posterior_analysis('conditional
decomposition',var_list_(i,:),M_.exo_names(j,:),Steps,options_,M_,oo_);

Error in ==> dynare_estimation_1 at 1081
oo_ = compute_moments_varendo('posterior',options_,M_,oo_,var_list_);

Error in ==> dynare_estimation at 62
dynare_estimation_1(var_list,varargin{:});

Error in ==> model1 at 120
dynare_estimation(var_list_);

Error in ==> dynare at 132
evalin('base',fname) ;

Here is the model and data files.

Hi Jean-Paul,

I think it is a bug that has been fixed in the unstable version. Please try with the unstable version and tells me how it goes.

Best

Michel