Re: st: ivreg2 question

[Jana von Stein <janavs@umich.edu>]

OK. All I've ever read is that it is "the convention" to include the  
exogenous predictors in step 1. I haven't looked far into the  
econometrics however. It makes no substantive theoretical sense for  
some of my step 2 predictors to be in step 1; that was my motivation...
On May 1, 2012, at 1:47 PM, Austin Nichols wrote:

> Jana von Stein <janavs@umich.edu>:
> No, because that would be econometrically illegitimate.
>
> On Tue, May 1, 2012 at 1:28 PM, Jana von Stein <janavs@umich.edu>  
> wrote:
> > Hello,
> >
> > Does anyone know if it is possible to prevent STATA from including  
> > in step 1
> > estimation the (exogenous) independent variables from step 2? From  
> > what I
> > can tell, partial doesn't do the trick.
> *
> *   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/

*
*   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/