survcomp (version 1.22.0)

ibsc.comp: Function to compare two IBSCs

Description

This function compares two integrated Briers scores (IBSC) through the estimation of the Brier scores (BSC) at some points in time. The statistical test is a Wilcoxon rank sum test for dependent samples.

Usage

ibsc.comp(bsc1, bsc2, time)

Arguments

bsc1

vector of BSCs computed from the first predicted probabilities for some points in time

bsc2

vector of BSCs computed from the second predicted probabilities for some points in time

time

vector of points in time for which the BSCs are computed

Value

p.value

p-value from the Wilcoxon rank sum test for the comparison ibsc1 < ibsc2

ibsc1

value of the IBSC for the first set of BSCs

ibsc2

value of the IBSC for the second set of BSCs

Details

The two vectors of BSCs must be computed from the same samples (and corresponding survival data) and for the same points in time. The function uses a Wilcoxon rank sum test for dependent samples.

References

Wilcoxon, F. (1945) "Individual comparisons by ranking methods", Biometrics Bulletin, 1, pages 80--83. Haibe-Kains, B. and Desmedt, C. and Sotiriou, C. and Bontempi, G. (2008) "A comparative study of survival models for breast cancer prognostication based on microarray data: does a single gene beat them all?", Bioinformatics, 24, 19, pages 2200--2208.

See Also

sbrier.score2proba, sbrier

Examples

set.seed(12345)
age <- rnorm(30, 50, 10)
size <- rexp(30,1)
stime <- rexp(30)
cens <- runif(30,.5,2)
sevent <- as.numeric(stime <= cens)
stime <- pmin(stime, cens)
##Brier scores
##size
dd <- data.frame("time"=stime, "event"=sevent, "score"=size)
bsc1 <- sbrier.score2proba(data.tr=dd, data.ts=dd, method="cox")
##size
dd <- data.frame("time"=stime, "event"=sevent, "score"=age)
bsc2 <- sbrier.score2proba(data.tr=dd, data.ts=dd, method="cox")
if(!all(bsc1$time == bsc2$time)) {
  stop("the two vector of BSCs must be computed for the same points in time!") }
ibsc.comp(bsc1=bsc1$bsc, bsc2=bsc2$bsc, time=bsc1$time)