MBonf.p.adjust {MHTdiscrete}

The adjusted p-values for Modified Bonferroni single-step FWER controlling procedure.

Description

The function for calculating the adjusted p-values based on original available p-values and all attaianble p-values

Usage

MBonf.p.adjust(p, p.set, alpha, make.decision)

Arguments

p

numeric vector of p-values (possibly with NAs). Any other R is coerced by as.numeric. Same as in p.adjust.

p.set

a list of numeric vectors, where each vector is the vector of all attainable p-values containing the available p-value for the corresponding hypothesis.

alpha

significant level used to compare with adjusted p-values to make decisions, the default value is 0.05.

make.decision

logical; if TRUE, then the output include the decision rules compared adjusted p-values with significant level \alpha

Value

A numeric vector of the adjusted p-values (of the same length as p) if make.decision = FALSE, or a list including original p-values, adjusted p-values and decision rules if make.decision = TRUE.

Note

The attainable p-value refers to the element of domain set of p-value for the corresponding hypothesis. For continuous test statistics, the p-value under true null are uniform distributed in (0,1), thus the p-values are attainable everywhere between 0 and 1. But for discrete test statistics, the p-value can only take finite values bewtween 0 and 1, that is the attainable p-values for discrete case are finite and countable, so we can assign them in a finite list p.set.

Author(s)

Yalin Zhu

References

Zhu, Y., & Guo, W. (2017). Familywise error rate controlling procedures for discrete data arXiv preprint arXiv:1711.08147.

See Also

Tarone.p.adjust, MixBonf.p.adjust, p.adjust.

Examples

p <- c(pbinom(1,8,0.5),pbinom(1,5,0.75),pbinom(1,6,0.6))
p.set <-list(pbinom(0:8,8,0.5),pbinom(0:5,5,0.75),pbinom(0:6,6,0.6))
MBonf.p.adjust(p,p.set)
## Compare with the traditional Bonferroni adjustment
p.adjust(p,method = "bonferroni")
## Compare with the Tarone adjustment
Tarone.p.adjust(p,p.set)