Statalist


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

st: noskip option with xtlogit


From   Garry Anderson <g.anderson@unimelb.edu.au>
To   statalist@hsphsun2.harvard.edu
Subject   st: noskip option with xtlogit
Date   Thu, 11 Sep 2008 19:02:30 +1000

Dear Statalist,

-xtlogit-  has a noskip option which, according to the documentation,
performs an overall likelihood-ratio test when either the re or fe
option is specified. It appears that the likelihood ratio test is always
performed when the fe option is specified, regardless of the noskip
option. The re option requires the noskip option to produce the LR
Chi2(#) statistic, and produces the Wald Chi2(#) statistic without the
noskip option.

Is there an option to produce the Wald Chi2(#) statistic when using the
fe option, or is this not applicable?

Secondly, my understanding is that the likelihood ratio statistic is
preferred to the Wald statistic, in which case would it not be
preferable to have the re option produce the LR Chi(#) statistic by
default, rather than specifying the noskip option? Maybe the default to
Wald Chi(#) for -xtlogit y x, re- is to be consistent with the p-values
in the body of the table produced by xtlogit, although -logit- produces
LR Chi(#) but Wald p-values in the body of the table.

Best wishes, Garry

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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