st: Re: "recursive var"

[Kit Baum <baum@bc.edu>]

What you are describing is not a VAR; it is a system of simuitaneous  
equations. A VAR is just what it says: an autoregression of each  
variable on the PAST values of itself and the other variables in the  
VAR. Once you have a current value of any of those variables on the  
RHS, you can no longer use OLS to estimate the equation. A VAR can be  
estimated with equation-by-equation OLS. What you describe requires  
instrumental variables techniques. If you want to estimate the entire  
system, use reg3 (three-stage least squares).

Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html


On Aug 27, 2005, at 2:33 AM, statalist-digest wrote:


> I am trying to run a recursive VAR (vector autoregression) in which  
> some
> equation(s) of the VAR system contains the current value of other
> variable(s) , not just past values of the variables.
>
> I  wonder if Stata allows for including some of the current values of
> variable(s).
> Anyone who is familiar with VAR in Stata, please reply.
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/