Error with bayesian mode_compute=6

General DSGE Modeling
1

Hi all,
I am estimate a model using bayesian method. So far, only mode_compute=6 can be estimated. After the bayesian IRF, i face the error as follow:

Index exceeds matrix dimensions.

Error in DsgeSmoother (line 80)
constant = SteadyState(bayestopt_.mfys);

Error in prior_posterior_statistics_core (line 173)
[alphahat,etahat,epsilonhat,alphatilde,SteadyState,trend_coeff,aK] = …

Error in prior_posterior_statistics (line 222)
[fout] = prior_posterior_statistics_core(localVars,1,B,0);

Error in dynare_estimation_1 (line 821)
prior_posterior_statistics(‘posterior’,dataset_);

Error in dynare_estimation (line 89)
dynare_estimation_1(var_list,dname);

Error in Ajello_AER16 (line 979)
dynare_estimation(var_list_);

Error in dynare (line 180)
evalin(‘base’,fname) ;

For the mod file, i have to skip one parameter estimation and for sg_ss, i use the mode that author provide as the init value and use the bound as 5%-95% in the paper. The only reason is that the estimation is freezing if i don’t do this. I use dynare 4.4.2 and matlab 2014a as author suggested. Can anyone have any idea? Thank you so much.

Ajello_AER16.mod (23.1 KB) Ajello_AER16_steadystate.m (2.2 KB) MYDATA_baseline.mat (19.4 KB) obj_dummy.m (57 Bytes) SS_Ajello_AER16.m (8.4 KB)

2

Sorry, but I have a really hard time working with that model due to some strange stuff going on in the steady state file. The steady state file seems to alter the parameter values in a non-deterministic way if the steady state cannot be computed. That will screw up estimation.

3

Thank you for your reply, the file is provided by the author so i don’t know if it is a good idea to modify it again. About the error, is it because the estimation is really sensitive to do the forecast ? it stop at the forecasting process.

4

The file as is will most probably return incorrect results and will produce cryptic crashes. Sorry, but there is not much one can do about it, except correct the file.