On Mon, 20 Aug 2007, Summers, Peter wrote:

> Do any of you have any experience using gretl for structural VAR 
> work? 

Structural VARs are not yet supported as such, although I think 
Jack Lucchetti wrote a gretl script that headed in that direction.  
Hopefully, Jack will be back on the list before long, and he can 
surely offer more help than I.

> On a related note, is there a way to get impulse responses from 
> a non-choleski decomposition? I'm thinking of something along 
> the lines of the "impulse" command in RATS: impulse(vcv=blah) 
> ... where 'blah' is my covariance matrix.

Not via a basic command, at this point.  However, you can get the 
companion matrix from a VAR, and you can get the contemporaneous 
covariance matrix, and you have a lot of matrix manipulation 
tools.

To get the companion matrix:

* estimate a VAR
* matrix A = $compan

(But in gretl prior to today's changes in CVS, and todays' Windows 
snapshot, "matrix A = $coeff" -- now $coeff gets you the regular 
coefficient matrix.)

To get the contemp covariance:

matrix V = $vcv

Allin Cottrell

Reply via email to