hierFDR.CF: Critical Function for all Hierarchical FDR Controlling...

In allenzhuaz/MHThier: Multiple Hypotheses Testing for Hierarchical Data

Description

Given some parameters of the descendant hypotheses and ancestor hypotheses, return the critical function using the generalized hierarchical FDR controlling procedure under various type of dependence.

Usage

hierFDR.CF(isTested, mi, li, l, depth, fdi, gdi, alpha, rOffset, type)

Arguments

isTested	logical; if TRUE (default), then the i-th hypotheses H_i will be tested; otherwise, H_i will not be tested.
mi	the cardinality of the set of descendant hypotheses of H_i.
li	the number of leaf hypotheses in the set of descendant hypotheses H_i.
l	the total number of leaf hypotheses.
depth	the cardinality of the set of ancestor hypotheses of H_i, referred to as the depth of H_i.
fdi	the cardinality of the set of all hypotheses with the given depth.
gdi	the cardinality of the union set of all hypotheses with all depth no deeper than the given depth.
alpha	the significant level used to calculate the critical values to make decisions.
rOffset	the offset increment for the number of rejections.
type	the type of dependence structure of the hierarchically ordered hypotheses. Currently, we provide four types of dependence: "positive", "arbitrary", "block positive" and "block arbitrary".

Value

A critical function of the index i and rejection number R.

Author(s)

Yalin Zhu

References

Lynch, G., Guo, W. (2016). On Procedures Controlling the FDR for Testing Hierarchically Ordered Hypotheses. arXiv preprint arXiv:1612.04467.

See Also

PositiveDeptCF, ArbitraryDeptCF, BlockPositiveCF, BlockArbitraryCF

allenzhuaz/MHThier documentation built on May 6, 2019, 1:30 a.m.