Re: [R] question about survSplit

From: Barker, Chris [SCIUS] <cbarker1_at_scius.jnj.com>
Date: Thu 18 May 2006 - 01:55:11 EST

Check the literature - an excellent article is "rhDNase as an example of recurrent Event Analysis, Therneau, Hamilton, Statistics in Medicine, Vol 16, 2029-2047" Implicitly the intervals are "( ... ]".

 -----Original Message-----
From: r-help-bounces@stat.math.ethz.ch
[mailto:r-help-bounces@stat.math.ethz.ch] On Behalf Of singyee ling

Sent:	Wednesday, May 17, 2006 8:01 AM
To:	r-help@stat.math.ethz.ch
Subject:	[R] question about survSplit

Dear R-users,

I use the survsplit function in the survival package to change my data into counting-process format

and the transformed format is as follow:

(a)

start stop event DP age ....

0      5         0        1       20
5     10        0        1       20
10    25       1        1       20


looking at the above three entries that belong to the same person, if an event happen at time 5, won't the person actually enter the risk set twice since there is another entry that start at time 5 and Cox proportional hazard model "won't know" that it actually belong to the same person.

Shouldn't it be like this?

(b)

start stop event DP age ....

0      5         0        1       20
6    10         0        1       20
11   25        1        1       20

or the R-function coxph has already take this into account when calculating its risk set although I believe (a) is actually the correct one. Your advice is greatly appreciated!

regards,
sing yee

        [[alternative HTML version deleted]]



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html

        [[alternative HTML version deleted]]



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Thu May 18 02:00:03 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Thu 18 May 2006 - 04:10:10 EST.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.