Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: gmm system using xtabond2


From   Roberto <terrymondo@onetel.net.uk>
To   statalist@hsphsun2.harvard.edu
Subject   st: gmm system using xtabond2
Date   Tue, 02 Nov 2004 12:26:20 +0000

Dear Statalist users
I am trying to replicate the gmm system DPD results with the xtabond2 module in Stata 8.

I have managed so far to replicate the gmm difference results estimating the model using the following syntax:

xtabond2 Y l.Y X1 X2 X3 year1-year11, gmm(Y, lag(2 2)) gmm(X1, lag(1 2)) gmm( X2 X3, lag(2 2)) iv(year1-year11, passthru) noleveleq robust small noc twostep

and the results perfectly match the output from DPD.

What I would like to do next is to estimate the gmm system. Say for example that I would like to include the first lag of the difference as an instrument for the level equation.

In the HELP for xtabond2 it is mentioned that “[…] The optional laglimits(a b) suboption can override these defaults: for the first-difference equations, lagged levels dated t-a to t-b are used as instruments, while for the levels equations, the first-difference dated t-a+1 is used”.
Unfortunately I have not been able to find a way to specify something like:
gmm(Y X1 X2 X3, lag(1,1) equation(level))
Thank you very much for your kind attention

Roberto Mura
PhD Candidate


*
* 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/




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index