Dynare forums

Posted: **Sun Nov 06, 2016 12:28 pm**

Hi Professor，
In dynare reference manuel, it is said that

After running estimation, the parameters M_.params and the variance matrix M_.Sigma_e of the shocks are set to the mode for maximum likelihood estimation or posterior mode computation without Metropolis iterations.
After estimation with Metropolis iterations (option mh_replic > 0 or option load_mh_file set) the parameters M_.params and the variance matrix M_.Sigma_e of the shocks are set to the posterior mean.

However, I want to get the estimation output

Code: Select all: oo_.PosteriorIRF.dsge.HPDinf/sup

where M_.params and M_.Sigma_e are set to the posterior mode ...Is that possible?

Or, is the estimation output

Code: Select all: oo_.PosteriorIRF.dsge.HPDinf/sup

the same no matter M_.params and M_.Sigma_e are set to the posterior mode or to the posterior mean?

Posted: **Sun Nov 06, 2016 6:13 pm**

The question is not well-defined.

Code: Select all: HPDinf/sup

is a "confidence band" across parameters. You cannot compute the variability of IRFS across parameters if you fix the parameters.

Dynare forums

Posterior Mode v.s. Posterior Mean

Posterior Mode v.s. Posterior Mean

Re: Posterior Mode v.s. Posterior Mean