Statalist The Stata Listserver


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

Re: st: RE: -clogit- on proportions data?


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: -clogit- on proportions data?
Date   Thu, 8 Mar 2007 13:09:55 -0500

Austin Frakt--
You should provide the reference for Green--do you mean Greene:
http://www.stata.com/bookstore/ea.html
and if so, what page number?
The model you described sounded like one where you have a set of
proportions y_ij that sum to 1 for each individual j, as in a y that
measures proportion of total assets in stocks, bonds, etc.  Then I
think you mean to estimate a fractional logit as in Papke and
Wooldridge (1996), see
http://www.stata.com/support/faqs/stat/logit.html
for the Stata command syntax, but with a seemingly unrelated
regression structure imposed on top, to account for the adding-up
constraint, and I thought Brian Poi's article and programs described
at
http://www.stata.com/statalist/archive/2004-01/msg00816.html
might be relevant.

On 3/8/07, frakt@bu.edu <frakt@bu.edu> wrote:
Thanks to Nick Cox and Austin Nichols (in private communication) for
suggestions on my post about -clogit-like estimation on proportions
data. It isn't clear to me that any of the Stata approaches are the
same as Green's / LIMDEPs. I don't know if I have the expertise to
confirm that or to appreciate the differences but I will look into it.

--Austin Frakt
*
*   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