Re: RE: RE: st: RE: GMM estimation: restricting parameter estimates

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

<>
On Jun 26, 2013, at 8:33 AM, Bley
 wrote:

> I did that but now, how do I get the standard errors and p-value for the variable v which is my variable of interest?
> Could you please tell me what procedure I can use to run a GMM including FIXED EFFECTS? I am not sure if the xtabond includes already fized effects.

help lincom

DPD estimators such as xtabond2 do not use fixed effects; they use first differencing, which like fixed effects remove the unobserved heterogeneity in a dynamic panel. But you do not seem to have a duynamic model.

I think you can easily estimate this model with IV-GMM and fixed effects with some simple algebra and Mark Schaffer's xtivreg2 from SSC:

g double yv = y - v
g double kv = k - v
g double hv = k - v
ssc inst ivreg2, replace
ssc inst ranktest, replace
ssc inst xtivreg2, replace
xtivreg2 yv a (kv hv = z1 z2 z3 z4 z5), fe gmm2s robust
lincom v = 1 - kv - hv

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
                                                                                                      http://www.crup.com.cn/Item/111779.aspx     


*
*   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/