ipcw: Estimation of censoring probabilities

Description

This function is used internally by the function pec to obtain inverse of the probability of censoring weights.

Usage

ipcw(formula,
     data,
     model = c("cox", "marginal", "nonpar", "aalen", "none"),
     times,
     otimes)

Arguments

formula

A survival formula like, Surv(time,status)~1, where as usual status=0 means censored. The status variable is internally reversed for estimation of censoring rather than survival probabilities. Some of the available models (see argument

data

The data used for fitting the censoring model

model

Censoring model used for estimation of the (conditional) censoring distribution.

times

Time values at which the probabilities of not being censored are returned

otimes

The time values, usually the original event times in the data.frame data, at which lagged individual probabilities of not being censored are returned.

Value

wtEstimated weights at times
wt.obsEstimated weights at individual time values otimes
fitThe fitted censoring model

Details

Inverse of the probability of censoring weights (IPCW) usually refer to the probabilities of not being censored at certain time points. These probabilities are also the values of the conditional survival function of the censoring time given covariates. The function ipcw estimates the conditional survival function of the censoring times and derives the weights. Currently the data set should be order by time to get the value wt.obs in the right order for some choices of model!

Examples

Run this code

dat=SimSurv(300)

dat <- dat[order(dat$time),]

# using the marginal Kaplan-Meier for the censoring times

WKM=ipcw(Hist(time,status)~X2,data=dat,model="marginal",times=sort(unique(dat$time)),otimes=dat$time)
plot(WKM$fit)
WKM$fit

# using the Cox model for the censoring times given edema

WCox=ipcw(Surv(time,status)~X2,data=dat,model="cox",times=sort(unique(dat$time)),otimes=dat$time)
WCox$fit

plot(WKM$fit)
lines(sort(unique(dat$time)),1-WCox$wt[1,],type="l",col=2,lty=3,lwd=3)
lines(sort(unique(dat$time)),1-WCox$wt[2,],type="l",col=3,lty=3,lwd=3)

# using the stratified Kaplan-Meier for the censoring times given X2

WKM2=ipcw(Surv(time,status)~X2,data=dat,model="nonpar",times=sort(unique(dat$time)),otimes=dat$time)
plot(WKM2$fit,add=TRUE)