[S] Cox PH Model

Mr J Gomez (ecrfk@csv.warwick.ac.uk)
Sat, 7 Nov 1998 15:13:26 +0000 (BST)


Dear splus users,

I am trying to estimate Cox Porportional Hazards Model with SPLUS. I have
multiple events per subject, varying covariates, discontinuous intervals
of risk and not all the subjects enter the risk set at time zero. I am
using Andersen and Gill method.

When I give SPLUS the following command:

coxph(Surv(t1,t2,depe) durac+expe+est+mujer+edad+pib+cluster(indiv),
data=agsplus)

where t1 is start time, t2 is stop time, depe is status, "durac, expe,
est, mujer, edad, pib" are covariates, indiv identifies every individual
and agsplus is the data file, I receive the following message:

Error in Surv(t1 , t2, depe): Stop time must be > start time Dumped.

I have checked for this in the data and all the starting times are
previous to the stopping ones, contrarily to what the error message seems
to suggest. I have already taken a look at the manuals, but I have not
been able to find anything wrong. Does anyone have any idea about what I
am missing?.

Thanks a lot for your help.

Jaime Gomez.
Department of Economics.
University of Warwick. UK.
(Visiting Research Student).
j.gomez@warwick.ac.uk

-----------------------------------------------------------------------
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