FDRcontrol: FDR controlling procedures

In xiangjiexue/npfixedcompR: Non-parametric estimation of mixing distribution with possibily fixed components

Description

Implementation of several FDR-controlling procedure in R.

Usage

adaptive.stepdown(pval, alpha = 0.05)

BH(pval, alpha = 0.05)

Arguments

pval	a vector of p-values (no necessarily sorted)
alpha	given FDR level

Details

The function adaptive.stepdown is a simple implementation of the adaptive step-down procedure described in Gavrilov et al. (2009)

The function BH is a direct implementation of the procedure described in Benjamini and Hochberg (1995).

Value

a list with num.rejection, the number of rejections computed by this function, and classifer, a vector of TRUE and FALSE; if TRUE, the corresponding input is regarded as null, and as non-null if otherwise.

Examples

adaptive.stepdown(pnorm(-abs(rnorm(1000, c(0, 2)))) * 2)
BH(pnorm(-abs(rnorm(1000, c(0, 2)))) * 2)

xiangjiexue/npfixedcompR documentation built on Jan. 1, 2021, 11:39 p.m.