st: Re: simultaneous equations with fixed effects and robust standard errors

[Kit Baum <baum@bc.edu>]

As Mark Schaffer's posts have often indicated, there is nothing  
stopping you from using a standard instrumental variables estimator  
in this context, in essence employing the linear probability model in  
the 'first stage' estimation. You need not use a probit nor logit to  
get consistent estimates of the equations of interest.


Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


On Jun 11, 2007, at 2:33 AM, Duy Hinh Khieu wrote:


> Something else comes up in my model in anticipation of arguments on  
> my research. I wonder if I can still use xtivreg2 if some dummies  
> in my model now become endogenous as follows:
>
> Eq 1: leverage = a0 + a1*maturity + a2*control variable (lots of  
> more control variable)
> Eq 2: maturity = b0 + b1*leverage + b2*Dummy endogenous (6 of them)  
> + b3*control variable
>
> The endogenous dummies are in equation 2 only. What I could think  
> of doing is to follow Maddala (1983) as I spelled out in an  
> biprobit thread. Briefly, I need to run probit of the dummies on  
> all exog var in equation 2, get the predicted values for the  
> dummies, plug the predicted values in eq. 2, and then use xtivreg2  
> to run eq.1 and eq.2 simultaneously.
*
*   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/