recurrent.marginal.mean: Estimates marginal mean of recurrent events

In timereg: Flexible Regression Models for Survival Data

Estimates marginal mean of recurrent events

Description

Fitting two aalen models for death and recurent events these are combined to prducte the estimator

\int_0^t S(u) dR(u)

the mean number of recurrent events, here

S(u)

is the probability of survival, and

dR(u)

is the probability of an event among survivors.

Usage

recurrent.marginal.mean(recurrent, death)

Arguments

recurrent	aalen model for recurrent events
death	aalen model for recurrent events

Details

IID versions used for Ghosh & Lin (2000) variance. See also mets package for quick version of this for large data mets:::recurrent.marginal, these two version should give the same when there are no ties.

Author(s)

Thomas Scheike

References

Ghosh and Lin (2002) Nonparametric Analysis of Recurrent events and death, Biometrics, 554–562.

Examples

### get some data using mets simulaitons 
library(mets)
data(base1cumhaz)
data(base4cumhaz)
data(drcumhaz)
dr <- drcumhaz
base1 <- base1cumhaz
base4 <- base4cumhaz
rr <- simRecurrent(100,base1,death.cumhaz=dr)
rr$x <- rnorm(nrow(rr)) 
rr$strata <- floor((rr$id-0.01)/50)
drename(rr) <- start+stop~entry+time

ar <- aalen(Surv(start,stop,status)~+1+cluster(id),data=rr,resample.iid=1
                                                     ,max.clust=NULL)
ad <- aalen(Surv(start,stop,death)~+1+cluster(id),data=rr,resample.iid=1,
                                                     ,max.clust=NULL)
mm <- recurrent.marginal.mean(ar,ad)
with(mm,plot(times,mu,type="s"))
with(mm,lines(times,mu+1.96*se.mu,type="s",lty=2))
with(mm,lines(times,mu-1.96*se.mu,type="s",lty=2))

timereg documentation built on Jan. 17, 2023, 5:11 p.m.