The individual effects are not estimated by xtabond2 (or any other DPD
system-GMM implementation).
It is true that the individual effects are present in the levels
equation. However, the orthogonality conditions for the levels equation
should only involve regressors that are orthogonal to the individual
effects either in level or in first-difference. In any case, the levels
equation does not allow you to estimate the individual effects.
Jean Salvati
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of
> Robert William Scott
> Sent: Friday, May 19, 2006 10:28 AM
> To: [email protected]
> Subject: st: Xtabond2 / System GMM Query
>
> Dear Statalisters,
>
> Having exhausted all available online sources of help, I
> appeal for the assistance of Stata users familiar with xtabond2.
>
> When estimating a system GMM for dynamic panel data using
> xtabond2, it is unclear/ambiguous whether the parameters of
> the country dummies/country-specific fixed effects should be
> estimated.
>
> Whilst country dummies are removed in the difference
> regression, they are included in the level regression.
>
> It is unclear whether their coefficients are calculated
> (automatically) by Stata in xtabond2, and if so, how may this
> data be extracted.
>
> Can anybody offer any assistance?
>
> Looking forward to any response!
>
> Yours faithfully,
> Robert
> ****************************
> Robert Scott, B.A., M.Phil
> Ph.D. Student
> Department of Economics
> Adam Smith Building
> University of Glasgow
> GLASGOW G12 8RT
> SCOTLAND
> ****************************
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
