Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: design issue with suspect endogeneity


From   "[email protected]" <[email protected]>
To   [email protected]
Subject   st: design issue with suspect endogeneity
Date   Mon, 24 Jan 2011 14:41:09 +0100 (CET)

Dear statalist,

I am sorry to pose the same question for the second time. I posted it on 
Friday night and I had no answer, and I´m afraid it was the time with the 
lowest visibility.

I have a research design doubt. 

I am considering the following model and the focus of the analysis is on a2:

y=a1*x+a2*(x*z)+a3*z+e

where z is a dummy variable

Economic theory also suggests that:

x=b1*z+v

If I find collinearity between x and z to be ok, can I estimate the first 
equation with OLS and disregard the second equation? 

My view is that I can safely use OLS only if cov(e,v)=0, but how can I test 
this condition?

Thanks a lot,


Andrea


*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index