Re: st: IV identification test using svy

[William Buchanan <william@williambuchanan.net>]

Hi Bill, 

You could run the model with ivreg2 (from SSC) with pweights.  The only major change you should notice in the models is that the standard errors in the non-svyset data will be larger.  I'm not positive of the effect it would have on the diagnostic tests that are available from -ivreg2- but it should at least help you to start moving in the right direction.

Billy



On Mar 3, 2012, at 12:36 PM, Bill Lin wrote:

> Dear All
> 
> Does any one know how to conduct over/under identification test for IV
> using the command: svy: ivregress.  I need to handle the complex
> survey design and want to use 2sls for IV estimation. ivreg2 does not
> work with the svy prefix. Thanks,
> 
> Bill
> 
> *
> *   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/