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

Re: st: Fixed effects ordinal probit regression


From   Chris Rohlfs <car@uchicago.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Fixed effects ordinal probit regression
Date   Fri, 5 Sep 2003 08:49:25 -0500 (CDT)

jim,

another possibility you might like to consider is the linear probability
model (or perhaps some ordered choice version).  while the LPM isn't
exactly the most elegant theoretically, it will produce consistent fixed
effects estimates -- and LPM generally produces very similar results to
probit & other binary choice estimation procedures.

random effects will solve the degrees-of-freedom problem -- i.e., that
your observations are not truly independent & so your estimates should
take the autocorrelation into account... random effects will not solve any
problems of omitted variables bias, however.

if you're using a fixed effects model to get around the problem of bias --
that is, that there's some time-invariant characteristic about the people
in your data (anything genetic, for instance) that's correlated with the
regressor of interest, then random effects won't help you -- but the
linear probability model will work.

chris

On Fri, 5 Sep 2003, Guillaume Frechette wrote:

> Hi Jim: as others have already hinted at, you cannot simply add dummies 
> (well you can but it doesn't provide a consistent estimator). This is 
> sometimes referred to as the incidental parameters problem. Alternatively 
> you can estimate a random effects ordered probit (findit reoprob or findit 
> gllamm).
> 
> Best,
> 
> Guillaume :-?
> 
> 
> >
> >Dear Sirs,
> >
> >I was wondering if there is such a thing as fixed effects ordinal probit
> >regression.  If so, could one simply add dummy variables for the panel
> >indicator (e.g., subject id) to the ordinal probit model to obtain fixed
> >effects estimates?  Also, when estimating a fixed effects regression model
> >with a subject-level effect, how problematic is it if there are missing
> >observations on the dependent variable for some subjects (i.e., unbalanced
> >panels)?
> >
> >Thank you very much.
> >
> >Best Regards,
> >
> >Jim Shaw
> >Graduate Research Associate
> >College of Pharmacy
> >The University of Arizona
> ><< Shaw,JamesW..vcf >>
> 
> _________________________________________________________________
> Send and receive larger attachments with Hotmail Extra Storage.   
> http://join.msn.com/?PAGE=features/es
> 
> *
> *   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/



© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index