sim.cif: Simulation of output from Cumulative incidence regression...

In mets: Analysis of Multivariate Event Times

Simulation of output from Cumulative incidence regression model

Description

Simulates data that looks like fit from fitted cumulative incidence model

Usage

sim.cif(cif,n,data=NULL,Z=NULL,drawZ=TRUE,cens=NULL,rrc=NULL,cumstart=c(0,0),...)

Arguments

cif	output form prop.odds.subdist or ccr (cmprsk), can also call invsubdist with with cumulative and linear predictor
n	number of simulations.
data	to extract covariates for simulations (draws from observed covariates).
Z	to use these covariates for simulation rather than drawing new ones.
drawZ	to random sample from Z or not
cens	specifies censoring model, if "is.matrix" then uses cumulative hazard given, if "is.scalar" then uses rate for exponential, and if not given then takes average rate of in simulated data from cox model.
rrc	possible vector of relative risk for cox-type censoring.
cumstart	to start cumulatives at time 0 in 0.
...	arguments for invsubdist

Author(s)

Thomas Scheike

Examples

data(bmt)

scif <-  cifreg(Event(time,cause)~tcell+platelet+age,data=bmt,cause=1,prop=NULL)
summary(scif)  
plot(scif)
################################################################
#  simulating several causes with specific cumulatives 
################################################################

cif1 <-  cifreg(Event(time,cause)~tcell+age,data=bmt,cause=1,prop=NULL)
cif2 <-  cifreg(Event(time,cause)~tcell+age,data=bmt,cause=2,prop=NULL)
# dd <- sim.cifsRestrict(list(cif1,cif2),200,data=bmt)
dd <- sim.cifs(list(cif1,cif2),200,data=bmt)
scif1 <-  cifreg(Event(time,cause)~tcell+age,data=dd,cause=1)
scif2 <-  cifreg(Event(time,cause)~tcell+age,data=dd,cause=2)
   
par(mfrow=c(1,2))   
plot(cif1); plot(scif1,add=TRUE,col=2)
plot(cif2); plot(scif2,add=TRUE,col=2)

mets documentation built on May 29, 2024, 3:51 a.m.