Page 1 of 1
Panel VAR
Posted: Sat Oct 17, 2009 10:45 pm
by bearlotus
Dear Tom:
I am wondering if there is any sample code on panel VAR? And how to incorporate panel VAR with sign restrictions identification methods? Would you please give some hint?
Thank you very much,
Lian
Re: Panel VAR
Posted: Mon Oct 19, 2009 9:06 am
by TomDoan
What panel VAR technique do you need? If you are assuming a homogenous representation, you can handle it like any other VAR.
Re: Panel VAR
Posted: Mon Oct 19, 2009 9:48 am
by bearlotus
TomDoan wrote:What panel VAR technique do you need? If you are assuming a homogenous representation, you can handle it like any other VAR.
Dear Tom:
I have small N (n=7), relatively larger T ( t>100, for most of the cases). but not all the country has uniform number of observations. Assuming homogenous representation, so I just set up the panel and then use Uhlig (2005) sign restriction methodology. Will that be ok?
Thank you very much for your guidance!
Re: Panel VAR
Posted: Mon Oct 19, 2009 1:08 pm
by TomDoan
If a homogeneous representation is OK, yes, you can do that. That might, however, be a big "if". Remember, that means not just the coefficients but the covariance matrix is the same from one cross section to another.