Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: long run restrictions in structural var (SVAR)


From   Fabian Irek <fabian.irek@student.uni-tuebingen.de>
To   statalist@hsphsun2.harvard.edu
Subject   st: long run restrictions in structural var (SVAR)
Date   Thu, 25 Feb 2010 17:18:40 +0100

I have a rather statistical question, but since I use STATA10 I thought it could be posted here.

I estimated a six - variable VAR and performed granger-causality tests. I would like to estimate a SVAR now, so I need 15 identifying restrictions (N squared minus N and both divided by 2). Additionally, some of my variables have a contemporaneous correlation of zero. My question now: Can I say that if the correlation of X and Y is zero AND if X does not Granger-cause Y that I have a possible long-run restriction (say that the sum of all coefficients,lagged and contemporaneous, of
X which are related to the contemporaneous Y is zero) then???

I know that both is not exactly, exactly the same, the Granger-causality is an overall F-Test which tests if a1=a2=a3=0 (where the a's are the coefficients on lagged parameters) whereas the long-run restriction says that the sum of the coefficients (both contemporaneous and lagged) is zero. But is it okay to argument that way?? At the end of the day, I wanna have identification because I am interested in the parameters of the remaining variables!

Thanks

Fabian


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index