Ekrem Kalkan <[email protected]> :
Read the manual entry on -gmm- in Stata 11 for a multi-equation IV
with different instruments by equation.
On Thu, Dec 10, 2009 at 2:43 PM, Ekrem Kalkan <[email protected]> wrote:
> Hello Dear Stata users,
>
> In a presentation prepared by C. Baum (see the site
> http://fmwww.bc.edu/ec-c/s2009/771/DIWMar2008.beamer.pdf, page 38),
>
> it is said that it is possible to restrict certain instruments to
> enter certain first-stage regressions by specifying a full
> simultaneous system of equations and by using the command "reg3". As
> far as I know in "reg3", all instruments specified in option exog( )
> are used in reduced form regressions. Does anybody know how to do
> Baum's suggestion ?
> More concretely, I am currently estimating a demand system using AIDS
> model. Prices (P1, P2 etc.) are assumed to be endogenous. I assume
> instrument IV1 is correlated with P1, but I do not want to use IV1 for
> P2, for example.
>
> Thank you.
>
> Ekrem Kalkan
*
* 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/
