CG.test {compound.Cox}

Testing survival difference of two groups via the CG estimators

Description

Testing survival difference of two prognostic groups separated by a prognostic index (PI). Survival probabilities are computed by the CG estimators (Yeh, et al. 2023).

Usage

CG.test(t.vec,d.vec,PI,cutoff=median(PI),alpha=2,
copula=CG.Clayton,S.plot=TRUE,N=10000,mark.time=TRUE)

Arguments

t.vec

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

d.vec

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

PI

Vector of real numbers (the values of a prognostic index)

cutoff

A number determining the cut-off value of a prognostic index

alpha

Copula parameter

copula

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

S.plot

If TRUE, the survival curve is displayed

N

The number of permutations

mark.time

If TRUE, then curves are marked at each censoring time

Details

Two-sample comparison based on estimated survival functions via copula-graphic estimators under dependent censoring. The D statistic (the mean vertical difference betewen two estimated survival functions) is used for testing the null hypothesis of no difference in survival. See Yeh et al.(2023) for details.

Value

test

Testing the difference of two survival functions

Good

Good prognostic group defined by PI<=c

Poor

Poor prognostic group defined by PI>c

Author(s)

Takeshi Emura, Pauline Baur

References

Emura T, Chen YH (2018). Analysis of Survival Data with Dependent Censoring, Copula-Based Approaches, JSS Research Series in Statistics, Springer, Singapore.

Rivest LP, Wells MT (2001). A Martingale Approach to the Copula-graphic Estimator for the Survival Function under Dependent Censoring, J Multivar Anal; 79: 138-55.

Yeh CT, Liao GY, Emura T (2023). Sensitivity analysis for survival prognostic prediction with gene selection: a copula method for dependent censoring, Biomedicines 11(3):797.

Examples

t.vec=c(1,3,5,4,7,8,10,13)
d.vec=c(1,0,0,1,1,0,1,0)
PI=c(8,7,6,5,4,3,2,1)

CG.test(t.vec,d.vec,PI,copula=CG.Clayton,alpha=18,N=100)
CG.test(t.vec,d.vec,PI,copula=CG.Gumbel,alpha=2,N=100)