Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: xtlogit with a fractional response


From   "Verkuilen, Jay" <JVerkuilen@gc.cuny.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: xtlogit with a fractional response
Date   Wed, 27 Feb 2008 15:14:30 -0500

> --- Maarten buis wrote:

>>Yes, I understood the fractional logit to mean the model proposed by
Papke and Wooldridge (1993). Such terminology does not tend to be
universal across disciplines, so I should have added the reference
(especially when I am making a claim about the history of statistics and
Nick is on watch).<< 

Mike Smithson and I have extended beta regression to the mixed
case---we're working on a paper on that but it's not in good enough
shape to send out yet. I don't have a Stata formulation for it, but it's
possible to do in SAS via NLMIXED and also using MCMC. I can't imagine
it would be that hard to port to Stata but I can't do it at present.
Based on several simulation studies the beta mixed model worked very
well. I have done several analyses of real data and it worked nicely on
that too. 

As to a history of the logit function in psychology, take a look at the
book: Stephen Link. (1992) The Wave Theory of Difference and Similarity.
LEA Press. 

Jay

*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index