Claude--
I'm sure someone can, but it is a tall order. Consider that an
augmented RE probit regression gives average effects for the
time-varying variables in an underlying FE model in this paper (see pp
21-24):
Wooldridge
"Simple Solutions To The Initial Conditions Problem In Dynamic,
Nonlinear Panel Data Models With Unobserved Heterogeneity"
http://cemmap.ifs.org.uk/wps/cwp0218.pdf
On 12/12/06, Claude Francoeur <[email protected]> wrote:
Hello,
Can anyone explain intuitively the difference between the random-effects
and the population-averaged xtprobit models and when each should be used?
And why the fixed-effects model is not appropriate with xtprobit?
Thank you in advance,
Claude
*
* 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/
