Re: RE: RE: st: ivreg2: interacting the endogenous regressor

[Christopher Baum <kit.baum@bc.edu>]

On May 26, 2012, at 8:33 AM, Vassilis wrote:

> I was concerned that I somehow needed to have a predicted "en" which,
> when multiplied by "ex", would be perfectly correlated with the
> prediction of "ex_en". But I see why this is wrong: in Wooldridge's
> terms, what I should be after is a "projection of the interaction"
> (which is what you proposed) and not the "interaction of the projection"
> (which is what I was after).

"I somehow needed to have a predicted "en"..."
Surely not. They are just separate regressors, and there is no reason, in the absence of degeneracy (e.g. ex a constant, and therefore useless) 
 to expect that projections of en and en*ex on a bunch of stuff will lead to predicted values that have any exact algebraic relationship.

Kit



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


*
*   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/