Statalist


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

RE: st: xtlogit with a fractional response


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: xtlogit with a fractional response
Date   Wed, 27 Feb 2008 11:17:42 -0000

What Maarten says in his last sentence presumably applies to a
particular kind of model. 

It is interesting (to me anyway) that the conditioned reflex to "logit"
among many analysts 
is that the root idea is using what we now call the logit function as a
link function for binary 
responses. That approach goes back at least as far as 1941.

Extensions e.g. to responses with multiple outcomes are then regarded as
extra tricks building on that. 

However, the logit idea has a much longer history for modelling growth
or decay in demography, 
ecology, physiology, chemistry etc. for problems with essentially
continuous responses. 
The point is elaborated in a graphical context in a note forthcoming in
Stata Journal 8(1). 

Nick 
n.j.cox@durham.ac.uk 

Maarten buis

--- Garry Anderson <g.anderson@unimelb.edu.au> wrote:
> I was wondering if there is a way that xtlogit can be used with
> fractional response panel data? 
> (Other options include glm and the cluster robust variance, or
> xtgee.)

-xtgee- relates to -glm- as -xtlogit- relates to -logit-. So 
-xtgee- seems to me the natural way to extend this to approach 
to panel data, especially since the fractional logit was 
originaly proposed as within the framework of quasi-likelihood. 


*
*   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