The RATS Software Forum

Dear Tom,
I'm not sure what kind of prior is the prior used in the bayesian var estimation by using the order
specify(type=symmetric,tightness= .2) .5
I read in a post that the defect prior in Rats is the normal Wishart prior.
Would you clarify this doubt?
Thank you very much for your attention.

Kadiyala and Karlsson refer to it as an "independent Normal" prior. The Normal-inverse Wishart will often give fairly similar results, but it requires Gibbs sampling as the sigma matrix isn't treated as fixed.

Kadiyala, K., & Karlsson, S. (1997). Numerical Methods for Estimation and Inference in Bayesian VAR-Models. Journal of Applied Econometrics, 12(2), 99-132.

Thanks you very much for your quickly answer Doan.