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


From   Scott Merryman <>
Subject   Re: st: REOPROB
Date   Wed, 4 Nov 2009 12:33:47 -0600

On Wed, Nov 4, 2009 at 12:08 PM, Tim Waring <> wrote:

> I need to fix both fixed and random effects within an ordered probit model.
>  REOPROB is the package for this, and I can't believe that it would NOT have
> the facility to fit FIXED effects as well as random ones, but I can't find
> documentation on how to fit fixed effects.

Sufficient statistics may not exist.  At least that is true for the
fixed-effects probit models.  From [XT] Page 410:

"xtprobit fits random-effects and population-averaged probit models.
There is no command for a
conditional fixed-effects model, as there does not exist a sufficient
statistic allowing the fixed effects
to be conditioned out of the likelihood. Unconditional fixed-effects
probit models may be fit with
the probit command with indicator variables for the panels. However,
unconditional fixed-effects
estimates are biased."

See also:


*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index