Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Log pseudo-likelihood


From   Richard Williams <Richard.A.Williams.5@ND.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Log pseudo-likelihood
Date   Fri, 28 May 2004 06:23:53 -0500

At 12:05 PM 5/28/2004 -0400, Marwan Khawaja wrote:
Dear All,
Sometimes the output from logit reports log-pseudo likelihood instead of
log-likelihood --
I do not know why -- Where can I find documentation of this?

I am using stata 8.2 on ms windows.

TIA, Marwan
You will see that when using robust standard errors (which are sometimes forced by the use of options such as cluster, or pweights). Are you encountering it in other cases? There is some discussion of this on p. 28 of the Stata 8 Survey Data Manual. It says that "pseudo-maximum likelihood methods" (which get used with robust SEs) are not "true likelihoods" and hence "standard LR tests are no longer valid".

Incidentally, there must have been some fine-tuning of this since the manuals came out. P. 305 of the Reference Manual G-M gives an example of logistic regression using robust standard errors. It reports the log likelihood; but when I replicate the example it reports the log pseudo-likelihood. There are also differences in the robust standard errors reported and in the Wald chi-square.



-------------------------------------------
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/




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index