compound.Cox (version 3.33)

surv.factorial: Factorial survival analysis under dependent censoring

Description

Perform factorial survival analysis under dependent censoring under an assumed copula (Emura et al. 2024).

Usage

surv.factorial(t.vec,d.vec,group,copula,alpha,R=1000,t.upper=min(tapply(t.vec,group,max)),
C=NULL,S.plot=TRUE,mark.time=FALSE)

Value

copula.parameter

Copula parameter

p

Estimates of treatment effects

Var

Variance estimates

F

F-statistic

c.simu

Critical value via the simulation method

c.anal

Critical value via the analytical method

P.value

P-value of the F-test

Arguments

t.vec

Vector of survival times (time to either death or censoring)

d.vec

Vector of censoring indicators, 1=death, 0=censoring

group

Vector of group indicators, 1, 2, ..., d

copula

Copula function: "CG.Clayton","CG.Gumbel" or "CG.Frank"

alpha

Copula parameter

R

The number of Monte Carlo simulations to find the critical value of the F-test

t.upper

Follow-up end (default is max(t.vec))

C

Contrast matrix

S.plot

If TRUE, the survival curve is displayed

mark.time

If TRUE, then curves are marked at each censoring time

Author

Takeshi Emura

Details

Estimates of treatment effects and the test results are shown.

References

Emura T, Ditzhaus M, Dobler D, Murotani K (2024), Factorial survival analysis for treatment effects under dependent censoring, Stat Methods Med Res 33(1):61-79.

Examples

library(survival)
data(cancer)
dat=subset(colon,etype==1)
## Treatment effects ##
t.vec=dat$time
d.vec=dat$status
trt=dat$rx
C12=matrix(c(1,-1,0),1,3,byrow=TRUE)
C13=matrix(c(1,0,-1),1,3,byrow=TRUE)
C23=matrix(c(0,1,-1),1,3,byrow=TRUE)
group=as.numeric(trt) # 1=Obs; 2=Lev; 3=Lev+5FU
# surv.factorial(t.vec,d.vec,group,alpha=2,copula=CG.Clayton)
# surv.factorial(t.vec,d.vec,group,alpha=2,copula=CG.Clayton,C=C12)
# surv.factorial(t.vec,d.vec,group,alpha=2,copula=CG.Clayton,C=C13)
# surv.factorial(t.vec,d.vec,group,alpha=2,copula=CG.Clayton,C=C23)
# alpha is a copula parameter