Error in dsge_simulated_theoretical_variance_decomposition

Dynare help (legacy posts)
1

Dear all,
I got the following message:

Index exceeds matrix dimensions.
Error in dsge_simulated_theoretical_variance_decomposition (line 113)
Decomposition_array(linea,(i-1)*nexo+j) = tmp{2}(i,j);
Error in posterior_analysis>job (line 61)
[nvar,vartan,NumberOfFiles] = …
Error in posterior_analysis (line 34)
oo_ = job(type,SampleSize,arg1,arg2,arg3,options_,M_,oo_);
Error in compute_moments_varendo (line 101)
oo_ =
posterior_analysis(‘decomposition’,var_list_(i,:),M_.exo_names(j,:),],options_,M_,oo_);
Error in dynare_estimation_1 (line 818)
oo_ =
compute_moments_varendo(‘posterior’,options_,M_,oo_,var_list_);
Error in dynare_estimation (line 89)
dynare_estimation_1(var_list,dname);
Error in banks_exp_K_b_est (line 1126)
dynare_estimation(var_list_);
Error in dynare (line 180)
evalin(‘base’,fname) ;

I can’t get which matrix has a problem.

Thank you in advance.
data_banks_new.m (5.72 KB)
banks_exp_K_b_est_steadystate.m (6.58 KB)
banks_exp_K_b_est.mod (18.5 KB)

2

This might be a bug related to the loading of previous draws. Could you try the most recent snapshot and report back. If it does not work, I would need the full directory in a zip-file, including the mode-file, the datafile and the mh-files.

3

Dear Johannes,

Thanks a lot your reply. The most recent snapshot doesn’t seem to be working too. I enclosed the full directory. Please, change the filename extension into zip.
banks_exp_K_b_est0.txt (607 KB)

4

Sorry, but the _results-file is missing.

5

Dear Johannes,

I’ve made a new run for the codes, but Dynare did not save the _results-file, perhaps, because of an error. However, without the option conditional_variance_decomposition, Dynare creates the _results-file. Information about the error is saved in log-file. I attached the zip-file of the full directory except for the _posterior_draws1-file as it is too big (58MB).

Thank you.
banks_exp_K_b_est0.zip (906 KB)

6

I got it. The problem is that all your variables have a unit root and thus infinite variance. No variance decomposition is possible in this case.

7

Thanks a lot!