RE: st: RE: Pseudo R2 for xtlogit

[Richard Williams <Richard.A.Williams.5@ND.edu>]

At 08:30 PM 8/8/2006, Scott Merryman wrote:
> Though the documentation states that e(ll_0) is saved, it is not returned
> after -xtlogit, re-
It always makes me a little nervous when I try to get something that 
Stata, in its wisdom, has chosen not to give me.  Is something 
omitted for sound statistical reasons far beyond my understanding, or 
is something omitted because nobody bothered to program it?  With 
that caution in mind, here is an apparent workaround:

. sysuse auto,clear
(1978 Automobile Data)

. qui xtlogit fore , i(rep) re

. local ll0 = e(ll)

. qui xtlogit fore mpg gear, i(rep) re

. local ll = e(ll)

. display "pseudo R2=" (`ll0' - `ll')/`ll0'
pseudo R2=.54613321

Basically, you are just estimating the null model yourself, something 
which Stata, for whatever reason, has not done itself.

Also, as Nick points out, there are many stats that go by the name 
Pseudo R^2.  For whatever reason, Stata has chosen to confer its 
official blessing on McFadden's R^2.  Long & Freese (2006) discuss 
others in their book, available from Stata Press.  My quick summary 
of these measures is at

http://www.nd.edu/~rwilliam/xsoc694/x05.pdf


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc 
*
*   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/