Re: RE: st: RE: RE: difference between robust and cluster option

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

<>
On Jul 20, 2010, at 2:33 AM, Jing wrote:

> Based on some expert's recommendation, I used "xtivreg2, cluster (firm
> year)" to cluster the standard errors both on firm and time. the results
> show that all my regressors are automatically included as instrumental
> variables. Could you please advise me that whether I need to define IV
> when run "xtivreg2, cluster (firm year)"?

This has nothing to do with the cluster() option (which you can see if you leave it out). xtivreg2 estimates IV models. A special case of that is standard fixed-effects models, in which all RHS variables serve as their own instruments (as they do in OLS). If you're using xtivreg2 in this context, it will automatically list them as instruments, but that is just semantic.

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/