st: gmm system using xtabond2

[Roberto <terrymondo@onetel.net.uk>]

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/