Statalist The Stata Listserver

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

RE: st: RE: Pseudo R2 for xtlogit

From   Richard Williams <>
To, <>
Subject   RE: st: RE: Pseudo R2 for xtlogit
Date   Tue, 08 Aug 2006 21:50:00 -0500

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

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):
WWW (department):
* For searches and help try:

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