Statalist


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

Re: st: SAS versus Stata, Panel Study Logit models


From   "David M. Drukker" <ddrukker@stata.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: SAS versus Stata, Panel Study Logit models
Date   Thu, 20 Dec 2007 10:54:56 -0600 (CST)

Richard Williams <Richard.A.Williams.5@ND.edu> asked about exactly matching the results from other software.

-xtgee- uses the sample size, N, as the divisor in calculating the
working-correlation matrix, as it should. Generalized-estimating-equations
estimators are sometimes implemented with ad-hoc small-sample adjustments
made to the divisors.

I do not know precisely what the other packages are doing, but specifying
the -nmp- option will change the divisor used in estimating the
working-correlation matrix from N to N-p, which is the most popular
small-sample adjustment.

-- David
ddrukker@stata.com

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