Linda Lei
2006-Aug-17  00:33 UTC
[R] putting the mark for censored time on 1-KM curve or competing risk curve
Hi All,
 
I'm trying to figure out the cumulative incidence curve in R in some
limited time. I found in package "cmprsk", the command
"plot.cuminc" can
get this curve. But I noticed that there is no mark for the censored
time there, comparing with the KM curve by "plot.survfit". Here are my
codes (attached is the data):
 
----------------
dat<-read.table("F://wendy/BMT data analysis/final
data.txt",header=TRUE,sep="\t")
 
library(cmprsk)
 
library(survival)
 
attach(dat)
 
par(mfrow=c(2,1))
 
curve<-cuminc(SURVEFP/365.25,STATREP,Ritux)
 
plot(curve,xlab="years",curvlab=c("No rituximab",
"Rituximab"),lty=1:2,col=c("red","blue"),mark.time=TRUE)
 
a<-survfit(Surv(SURVEFP/365.25,STATREP)~Ritux,type="kaplan-meier")
 
plot(a,main="OS for whole
group",conf.int=F,xlab="years",ylab="probability",lty=1:2,col=c("red","b
lue"))
-------------------
 
Could you help me with how to put the mark for censored time on
cumulative incidence curve, or maybe how to get the 1-KM curve with the
mark for censored time? This will help me a lot!
 
Thank you very much!
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: final data.txt
Url:
https://stat.ethz.ch/pipermail/r-help/attachments/20060816/6b2d2754/attachment.txt
Linda Lei
2006-Aug-23  18:49 UTC
[R] putting the mark for censored time on 1-KM curve or competing risk curve
Hi All,
 
I'm trying to figure out the cumulative incidence curve in R in some
limited time. I found in package "cmprsk", the command
"plot.cuminc" can
get this curve. But I noticed that there is no mark for the censored
time there, comparing with the KM curve by "plot.survfit". Here are my
codes (attached is the data):
 
----------------
dat<-read.table("F://wendy/BMT data analysis/final
data.txt",header=TRUE,sep="\t")
 
library(cmprsk)
 
library(survival)
 
attach(dat)
 
par(mfrow=c(2,1))
 
curve<-cuminc(SURVEFP/365.25,STATREP,Ritux)
 
plot(curve,xlab="years",curvlab=c("No rituximab",
"Rituximab"),lty=1:2,col=c("red","blue"),mark.time=TRUE)
 
a<-survfit(Surv(SURVEFP/365.25,STATREP)~Ritux,type="kaplan-meier")
 
plot(a,main="OS for whole
group",conf.int=F,xlab="years",ylab="probability",lty=1:2,col=c("red","b
lue"))
-------------------
 
Could you help me with how to put the mark for censored time on
cumulative incidence curve, or maybe how to get the 1-KM curve with the
mark for censored time? This will help me a lot!
 
Thank you very much!
 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: final data.txt
Url:
https://stat.ethz.ch/pipermail/r-help/attachments/20060823/6e76f3b5/attachment.txt