[18812] in s-news-athena
[S] Coxph log likelihood
daemon@ATHENA.MIT.EDU (Therneau, Terry M., Ph.D.)
Mon Aug 2 11:23:06 1999
Date: Mon, 2 Aug 1999 10:13:24 -0500
From: "Therneau, Terry M., Ph.D." <therneau@mayo.edu> (Terry Therneau)
Message-Id: <199908021513.KAA21400@rocky.Mayo.EDU>
To: ellis@neuron.cpmc.columbia.edu
Cc: s-news@wubios.wustl.edu
Steve Ellis writes:
" ...and it appears that the "loglik" values returned
as part of the object returned by coxph are off by a factor of 2... "
The function is correct, but I would guess that your simulation isn't.
Did you remember that it is 2*(difference in loglik) that has a chisquare
distribution?
Terry M. Therneau, Ph.D. (507) 284-3694
Head, Section of Biostatistics (507) 284-9542 FAX
Mayo Clinic therneau.terry@mayo.edu
Rochester, Minn 55905
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
