s-news
[Top] [All Lists]

Re: [S] Generalized additive models for matched case-control data?

To: peterc@u.washington.edu
Subject: Re: [S] Generalized additive models for matched case-control data?
From: "Therneau, Terry M., Ph.D." <therneau@mayo.edu> (Terry Therneau)
Date: Mon, 28 Jun 1999 09:23:52 -0500
Cc: S-news@wubios.wustl.edu
Sender: owner-s-news@wubios.wustl.edu

  1. Case-control logistic can be done with coxph:
        status: 1=case, 0=control
        time:  dummy value, any positive number, I usually set it to 1.
        group: case-control group

        coxph(Surv(time, status) ~ x1 + x2 + ... + strata(group), 
                data=mydata, method='exact')

 This uses the "exact partial likelihood", which is numerically identical
to the likelihood for a matched case-control logistic.

  2. If you have the latest survival routines ("survival5"), then you can
use 'pspline(age, df=3)" to fit a smoothing spline on age.

        Terry Therneau
-----------------------------------------------------------------------
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

<Prev in Thread] Current Thread [Next in Thread>