Obtain Bayesian IRFs with 68% credibility intervals

Estimation ML/Bayesian estimation
1

Hi Everyone,

Do you how whether there is a way to obtain the 68% credibility intervals for Bayesian IRFs before or after estimation?

I know that I can access the posterior IRFs through the oo_.PosteriorIRF.dsge:

  • deciles which contains the 9 deciles for the posterior distribution of this IRF
  • HPDinf the 5%
  • HPDsup the 95%

I guess I can change the mh_conf_sig, but the issue is that I want to keep the 90% for the model parameters but take the 68% for the impulse responses.

Thanks,
Kostas

2

I think I found a way after estimation.

options_.mh_conf_sig = 0.84
PosteriorIRF('posterior')
3

Yes, that should work.

1 Like