Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: 2SLS with multiple endogenous variables


From   [email protected]
To   [email protected]
Subject   Re: st: 2SLS with multiple endogenous variables
Date   Sat, 15 Dec 2007 19:08:04 +0100

I suppose -cmp- from ssc may also help

Nicola

At 02.33 14/12/2007 -0500, you wrote:
Ana said

I am trying to estimate a 2SLS model with two
endogenous variables (y2, y3) which have different
exogenous explanatory variables.  That is,
y2=f(z1,z2) and y3=f(z3,z4,z5).  Is there any Stata
command that will allow the estimation of this model?


No, this is a FAQ. 2SLS requires that you regress each endog on all  
of the instruments, included and exlcuded.

If you specify a full mulitple-equation model (e.g. 3SLS) you can  
specify the exclusion restrictions that apply to the RHS endogenous  
variables (as they are then LHS in each of  their own equations) but  
in a 2SLS/IV context, it cannot be done. Consistent estimation of  
2SLS requires that all instruments appear in each 'first stage'  
regression.

See the Stata FAQs.

*
*   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index