st: 2SLS IV estimation CODE

[David Torres <writeon4truth2@msn.com>]

Hi,

So I've been all over the web, the Stata Journal, and a few Stata books and I still haven't been able to find any code useful for building a three-level random coefficients model that can take advantage of the two-stage least squares instrumental variables estimator or similar.

I'd very much appreciate the help of a Stata pro to help me construct such a model.

As previously stated, my data are repeated measures (maximum of five) of students nested within schools.  The endogenous variable I'm instrumenting for is a binary indicator for whether (1) or not (0) parents exercised choice to send their child(ren) to a school outside the zoned area.  Controls include typical factors like race, gender, and poverty status at level 2, and school type (regular public, school-within-a-school, school-wide magnet) at level 3.

level-1: intercept year
level-2: intercept black hispan frlunch poverty
level-3: intercept swas swp

Anyone? Anyone?

I'm also open to suggestions for other approaches or data setup that might work as well.

David DIEGO Torres 		 	   		  
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/