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: Treating interactions exogenously or endogenously in xtabond2
From
john ebireri <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: Treating interactions exogenously or endogenously in xtabond2
Date
Mon, 29 Oct 2012 07:57:05 -0700 (PDT)
Hello,
I have a model where i have created interaction terms from the endogenous variables (banking market structures)in the gmm statement and industrial characteristics. I am not sure of how to treat these interactions. In other words, should i treat them endogenously (include them in the gmm statement) or exogenously (include them in the iv statement) ?
Secondly, when i treat the interaction terms endogenously, i hardly get any significant interactions from the results in all the regressions. But when i treat the interaction terms exogenously, i have all the interaction terms to be significant but with the same signs for all the regressions. in other words, the endogenous variables affects all the industrial characteristics under consideration in the same way. This seems strange to me. can i get any advice on this?
An example of one of the models is stated below:
xtabond2 vagr l.vagr bcon bcXed fbank fbXed gbank gbXed li finan_dev3 gdpgr inf trade yr1996 yr1997 yr1998 yr1999 yr2000 yr2001 yr2002 yr2003 yr2004 yr2005 yr2006, gmmstyle (l.vagr bcon fbank gbank bcXed fbXed gbXed, laglimits (2 1) collapse) ivstyle (li finan_dev3 gdppcgr inf trade yr1996 yr1997 yr1998 yr1999 yr2000 yr2001 yr2002 yr2003 yr2004 yr2005 yr2006, equation (level)) twostep robust small
Thanks.
John.
*
* 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/