[R] kaplan-meier error

[Marc Schwartz]

On Jul 21, 2009, at 7:23 PM, Joseph Magagnoli wrote:

> Hi all,
> I am getting an error in my code and I don't know what the problem is.
> I am using R 2.9 on ubuntu.   my code is as follows:
>
> ## Libraries ##
> library(survival)
> library(foreign)
>
> ## reading data ##
>
> data<-read.dta("http://psfaculty.ucdavis.edu/bsjjones/cabinet.dta")
> head(data)
> attach(data)
>
>
> fit1<-survfit(Surv(durat,censor))
> and I get the following error
>> fit1<-survfit(Surv(durat,censor))
> Error in survfit(Surv(durat, censor)) :
>  Survfit requires a formula or a coxph fit as the first argument
> any help is greatly appreciated
> Joe


Try:

   fit1 <- survfit(Surv(durat, censor) ~ 1, data = data)

See ?survfit and note in the description of 'formula':

   For a single survival curve the right hand side should be ~ 1.

I would also avoid using attach() as it can create subsequent  
problems, which is why there is the 'data' argument to most R  
functions that take a formula argument.

HTH,

Marc Schwartz