Dear Statalisters,
Would someone be able to explain in plain language why -xtnbreg- with
the "conditional fixed effects" option (-xtnbreg, fe-) drops from the
estimation those i (groups) for whom the dependent (count) variable is
always zero, while -xtnbreg- with a random-effects estimator does not
drop any of those same i (groups)?
(I am aware that for -xtnbreg- "fixed effects" and "random effects" mean
something rather different from the OLS context -- here they refer to
how values for alpha, the over-dispersion parameter, are allocated. For
both the RE and conditional FE estimators, alpha is constant within
groups).
Much thanks for any explanations...
Jason Webb Yackee, PhD Candidate; J.D.
Fellow, Gould School of Law
University of Southern California
jyackee@law.usc.edu
Cell: 919-358-3040
*
* 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/
