The model I am trying to estimate is the RE probit model. I find that as soon as I include the lagged dependent variable as a regressor, the program does not move away from the initial ordinary probit estimates. This is the case even if the 'rho' is 0.4 as limdep estimates. Changing the number of quad points to 20 in stata also does not help. Limdep has no problem in locating an estimate for 'rho' with a well defined std. error!
Hope someone can help me with this problem.
best
wiji
=================================
Professor Wiji Arulampalam,
Department of Economics,
University of Warwick,
Coventry,
CV4 7AL,
UK.
Tel: +44 (24) 7652 3471
Sec. Tel: +44 (24) 7652 3202
Fax: +44 (24) 7652 3032
email: [email protected]
http://www.warwick.ac.uk/Economics/arulampalam/
RES2003: http://www.warwick.ac.uk/res2003/
>>> [email protected] 09/16/02 02:35PM >>>
"Wiji Arulampalam" <[email protected]> wrote:
>
>Dear Stephen,
>In most of the runs I get the result that there is no unobserved
>heterogeneity as the coefficient estimates are identical to ordinary probit
>when I use xtprobit. xtprobit also does not give me a std error for the
>rho. But I have a mle for RE probit in stata and I have just run that and
>find that my results are exactly the same as Limdep's.
>I wonder what is happening?
I find this very interesting and I hope someone will have something more
enlightening than me to say about this. I have inquired the list in the past
about a similar problem but got no answer. However, the difference between
your results and the one from xtprobit could simply be due to the starting
value. I also think Stephen's point was a good one, I have sometimes had
results without std errors but the problem went away once I changed the
number of quad. points. Bellow you'll find an answer I received from Stata
to a related but different question (something like "what does it mean when
Stata gives me missing values for the std. error after estimating a logit
with the cluster option"). Nonetheles, I would like to know what is the
source of the problem since Limdep does find a better solution, although the
log likelihood is not that much different.
g
_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com
*
* 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/
*
* 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/
