Re: st: STATA CODE INFO.: ivreg2 & ivregress

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

<>
On Jul 19, 2010, at 2:33 AM, Bond wrote:

> I would like to know whether there is any difference between the following codes
> or both the codes perform same task.......
> 
> (1) ....ivreg2, gmm2s and ....ivregress, gmm.
> (2) ....ivreg2, cue and ......ivregress, igmm
> 
> I think both the codes in (1) are performing the same task of instrumental
> variable two-step GMM estimation and codes in (2) are performing the same task
> of instrumental variable iterative GMM estimation.
> 
> Please correct me, if I am wrong.

Re (1): you are right.
Re (2): you are wrong. GMM-CUE of Hansen, Heaton, Yaron is not the same iterated GMM estimator as that included in official -ivregress-. They are both valid estimators, but they use different criteria for the iterative technique.

Kit
coauthor of -ivreg2-



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/