Re: st: Windmeijer's corrected se in xtivreg2 gmm2s

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

<>
Thank you Anees for your help. I have read all the relevant literature in detail. I have used xtabond2 for my dataset, now I want to analyse the subsample for something. The problem is the number of observations is 156 in the sub sample and I cant apply xtabond2 because of small sample size. I have applied xtivreg2 and now I want to correct the estimates for small sample size bias (Windmiejer's correction).

I am interested in knowing how can I apply the Windmeijer's correction on xtiverg2 gmm2s cluster(id).


The Windmeijer correction is for Arellano-Bond and Blundell-Bond "DPD" models. xtivreg2 by Mark Schaffer does not estimate those DPD models. It would not be appropriate to apply xtivreg2 with fixed effects to a dynamic panel model. The fact that xtivreg2 (and the underlying ivreg2 of Baum-Schaffer-Stillman) has a GMM option does not mean that it can consistently estimate dynamic panel models.

When you say you have only 156 obs. in the subsample, what is N and what is the average T?

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/