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

st: The use of (time) dummies in a dynamic system GMM framework, usingxtabond2


From   Joost Vandewege <joost.vandewege@ugent.be>
To   statalist@hsphsun2.harvard.edu
Subject   st: The use of (time) dummies in a dynamic system GMM framework, usingxtabond2
Date   Fri, 20 Feb 2004 14:18:59 +0100

Dear stata-user,

Can you tell me what is the right way to include time dummies, or
dummies in general when using the xtabond2 command?  The difference
equation cannot deal with normal dummies, while the level equation does
need normal (0-1) dummies. I have tried to replicate DPD98-output (where
there is an option to 'use time dummies') with identical datasets by
including the dummies as they are (0 or 1)for the level equation, and by
including a transformed dummy series (that becomes the normal dummy
series after taking first differences) for the difference equation.
However, this does not lead to the required results, and replication of
DPD-coŽfficients seems impossible.  I know that standard errors are not
comparable between stata and DPD, because Stata uses Windmeijer
correction, but the replication of the coefficients themselves ought to
be possible.  However, I can't seem to succeed.

Thus; in one sentence: how does one correctly enter dummy variables with
the xtabond2 command? What should those dummies look like?

Kind Regards,

Joost.

--
Joost vandewege
Department of Social Economics
Faculty of Economics and Business Administration
Hoveniersberg 24
9000 Gent
Belgium
phone: +32 (0)92647877
joost.vandewege@UGent.be



*
*   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