[R] Creating a survival object with and without an event indicator

wwreith reith_william at bah.com
Thu May 3 14:57:56 CEST 2012


When reading about surv(). I saw the following statement "Although unusual,
the event indicator can be omitted, in which case all subjects are assumed
to have an event." 

So I tried the following

1. survobj<-surv(mydata$Time) vs.  2. survobj<-surv(mydata$Time,
mydata$Event) 

where mydata$Event is a column of all ones. I did not get the same answer
when I ran survreg(survobj~shift, data=mydata) on each. The second case
actually failed to converge. I then tried a column of all zeros just to make
sure I wasn't misreading something but still got the warning message failed
to converge. 

Can anyone explain what exactly R is doing  for case 1?

--
View this message in context: http://r.789695.n4.nabble.com/Creating-a-survival-object-with-and-without-an-event-indicator-tp4605945.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list