permGS (version 0.2.5)

permLR: Convenience function which calls createPermGS and nextStage to perform fixed sample size permutation test without imputation

Description

Convenience function which calls createPermGS and nextStage to perform fixed sample size permutation test without imputation

Usage

permLR(formula, data, B = 1000, alpha = 0.05, pool = TRUE,
  type = c("logrank", "Gehan-Breslow", "Tarone-Ware", "Prentice",
  "Prentice-Marek", "Andersen-Borgan-Gill-Keiding", "Fleming-Harrington",
  "Self"))

Arguments

formula

a formula object, as used by coxph, left hand side must be a 'Surv' object, right hand side must only consist of a factor (treatment indicator) and optionally a special strata() term identifying the permutation strata

data

a data.frame or list containing the variables in "formula", by default "formula" is evaluated in the parent frame

B

number of random permutations (default: 1000)

alpha

significance level (default: 0.05)

pool

if TRUE impute event times from Kaplan-Meier estimator calculated from pooled data

type

logrank weights to be used with coin::logrank_trafo

Value

An object of class permGS

Examples

# NOT RUN {
## Two-sided permutation test
T <- rexp(100) ## event times
Z <- rbinom(100, 1, 0.5)  ## treatment assignment
C <- rexp(100) ## drop-out times
data <- data.frame(time=pmin(T,C), status=T<=C, Z=Z)
x <- permLR(Surv(time, status) ~ Z, data, alpha=c(0.025, 0.025))
summary(x)
# }