Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: system GMM


From   "Rubil Ivica" <[email protected]>
To   <[email protected]>
Subject   st: system GMM
Date   Fri, 6 Sep 2013 13:32:03 +0200

Dear all,

I would like to use the Arellano-Bover/Blundell-Bond system GMM
estimator via -xtabond2- because (a) I do not have any available
instruments outside of the data, (b) I want to be able to test
overidentifying restrictions, (c) there are individual fixed effects,
(d) I have a "large N, small T" panel. But, I do not want to include the
lagged dependent variable on the right-hand side. In other words, I want
my specification to be:

level equation: y(i,t)=
alpha+beta*regressor_of_interest(i,t)+gama*controls(i,t)+fixed_effect(i)
+e(i,t)
diff. equation:
y(i,t)-y(i,t-1)=beta*[regressor_of_interest(i,t)-regressor_of_interest(i
,t-1)]+gama*[controls(i,t)-controls(i,t-1)]+[e(i,t)-e(i,t-1)] 

Does this specification make any sense without the lagged y?

Thanks.

ivica 

*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index