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: Manipulating SVAR estimates for counterfactual analysis


From   Sören Karau <soerenkarau@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Manipulating SVAR estimates for counterfactual analysis
Date   Wed, 20 Apr 2011 14:35:19 +0100

Dear Statalist,

I am about to conduct counterfactual simulations after using a SVAR model.
In order to do so, I use a usual Cholesky decomposition and constrain the A
and B matrices like the following (assuming 4 variables in the system):

matrix A =3D (1,0,0,0\.,1,0,0\.,.,1,0\.,.,.,1)
matrix B =3D (.,0,0,0\0,.,0,0\0,0,.,0\0,0,0,.)

I let stata estimate the coefficients.

In a second step, I need to manipulate those coefficients in order to
simuluate what would happen to the system, if a certain variable did not
react to shocks from other variables, while leaving all other estimates as
before. If I understood correctly what I read, one can do that by imposing
that the cross-correlation between the variables under consideration is set
to zero. For instance, if it was variable 4 I was interested in, I would
need to change the estimated coefficients of A_[1,4], A_[2,4], A_[3,4] to
zero (for all lags), the rest remains as estimated before.

How can I do that in stata? I have stored the estimates as a variable, but
how can I directly impose that for an impulse reponse analysis stata should
use the estimated matrix using my additional restrictions
(A_[1,4]=3DA_[2,4]=3DA_[3,4]=3D0)?

I would greatly appreciate any help, thank you so much in advance!

Soeren Karau
Soerenkarau@gmail.com
*
*   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