Richard Williams wrote:
At 03:43 PM 12/19/2007, Richard Williams wrote:
Furthermore, I can also get SPSS to EXACTLY reproduce my Stata
results. You use the adjustcorr=no option on SPSS GENLIN to do this...
Alas, this does not mean anything to me, but apparently this is
something SAS and SPSS think should be done by default whereas Stata
does not. Who is right? And if I were so inclined, is there an
option I could use in Stata that would do this adjustment?
One more tidbit: digging around the SPSS documentation, I found
"By default, the procedure will adjust the correlation estimates by
the number of nonredundant parameters. Removing this adjustment may
be desirable if you want the estimates to be invariant to
subject-level replication changes in the data. "
This still doesn't mean much to me, but apparently Stata thinks it is
better for the estimates to be invariant to subject-level replication
changes in the data? But if I disagree with Stata and want to do it
like SAS and SPSS do by default, am I out of luck?
--------------------------------------------------------------------------------
Not sure, but nothing jumps out of Stata's on-line help file about this.
Could you have hit upon what's behind the statment, "There is one difference
between SAS 6.12 TS Level 0060 (and Windows version 4.10.1650) and Stata
that cannot be explained: The results (beta, working correlation matrix, and
standard errors) of using PROC GENMOD do not match xtgee when panels are
unbalanced." in www.stata.com/support/faqs/stat/xtgeesas.html ?
Joseph Coveney
*
* 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/
