Coviance of asymmetric Bivariate-GARCH models

Discussions of ARCH, GARCH, and related models
wendyyuan
Posts: 14
Joined: Wed Jun 17, 2009 6:07 am

Coviance of asymmetric Bivariate-GARCH models

Unread post by wendyyuan »

Hey, Everyone,
I am programming asymmetric BI-GARCH models, such as GARCH-GJR, EGARCH, TGARCH. When we use Maximize command, we have to define the covariance formula. Anybody knows where can i find covariance equations?

Thanks a lot in advance
TomDoan
Posts: 7814
Joined: Wed Nov 01, 2006 4:36 pm

Re: Coviance of asymmetric Bivariate-GARCH models

Unread post by TomDoan »

You shouldn't be trying to use MAXIMIZE unless you have a model that you can't estimate using GARCH. Nothing you're describing appears to be outside the abilities of the GARCH instruction. I would note that you're also listing univariate models. EGARCH, for instance, isn't a well-defined bivariate procedure since it describes only the method used for the univariate variances. There are several ways to combine those into a covariance matrix (CC and DCC).
Post Reply