st: RE: Calculating elasticities of an endogenous covariate with interaction terms

[Jason Russ <jasondruss@gmail.com>]

Hi Eilya,

Thank you very much for your reply.

I have tried the ivreg command, and it will not let me run an IV
regression with an endogenous interaction term.

If I were to calculate the two stages separately, then the standard
errors in the second stage would not be adjusted, and would therefore
be inaccurate. Then when I use the margins command to calculate the
elasticities, those standard errors would also be inaccurate, I
presume (since the standard errors from the 2sls regression are used
to calculate them).

Would bootstrapping then be the best way to calculate efficient
standard errors for the elasticities? Or is there some way to correct
the standard errors in the second stage before calculating the
elasticities?

Thanks again,
Jason
*
*   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/