t2p: Permutation p-value

In floatofmath/adaperm: Permutation tests for adaptive designs

Description

Permutation test p-value

Usage

t2p(t, dist, type = c("non-randomized", "midp", "randomized",
  "davison_hinkley"))

Arguments

t	observed test statistic
dist	permutation distribution
type	type of p-value to compute

Details

Type "non-randomized" will give the standard p-value, which is define by the proportion of sample permutations with larger or equal test statistic than the observed one. "midp" will give the mid-p value which is the proportion of sample permutations with larger test statistic plus one half the proportion of permutations with equal test statistic. "randomized" will return the proportion of permutations with larger test statistic plus a random number that is uniformly distributed between zero and the proportion of permutations with test statistic equal to the observed.

Value

p-value

Author(s)

Florian Klinglmueller

floatofmath/adaperm documentation built on May 16, 2019, 1:18 p.m.