Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

st: weak IV tests using xtabond2


From   David Parsley <David.Parsley@owen.vanderbilt.edu>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: weak IV tests using xtabond2
Date   Wed, 20 Mar 2013 21:52:42 +0000

I am using xtabond2 (stata version 11.2) and would like to examine 1st stage regressions and test statistics. However, xtabond2 doesn't seem to report first stage regressions. Can someone point me to an explanation (online somewhere?) of how I can produce exactly the 1st stage regressions xtabond2 is performing?  Then, I guess I could use something like xi: ivreg2?

I am also looking for a (presumably) standard weak IV test as output from xtabond2 and seem not to be able to find it.  I have read the "How to do xtabond2:..." from the Stata Journal (2009) as well as other general references, but have struck out so far.

I have a panel with > 1000 cross-sections, and potentially 100 or so observations, but with missing values some members of the panel have very few observations (< 10). My STATA command looks something like:

xi:xtabond2 y x1 x2 x3 x4 l.y i.year, orthogonal  twostep robust iv(x5 x6 x7 x8) gmm(y  x3 x4)  lag(3 4))

where x5-x8 are my exogenous instruments, and I limit the lags to mitigate instrument proliferation.

David Parsley

*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index