pValueAfterMultipleTesting: Multiple testing correction procedure for ncpc()

In ddgraph: Distinguish direct and indirect interactions with Graphical Modelling

Description

This function is only for DDGraph with multiple testing correction enabled. The overall procedure is similar to that described in (Li&Wang 2009). This is a helper function for DDDataSet:ncpc(). The single P-value of D-separation is substituted in the list of P-values, P-values adjusted and the resulting P-value after correction in the context of other P-values reported.

Usage

pValueAfterMultipleTesting(dsep, x, adjC.pvals.at.n, p.value.adjust.method)

Arguments

dsep	the conditional independence test result (of type CITestResult)
x	the index of the variables
adjC.pvals.at.n	the p values associated with the variables at size n of conditioning set (list [[n]] -> [pvals])
p.value.adjust.method	the p value adjustment method (same as in p.adjust())

Value

the p value after multiple test correction (if any)

References

J. Li and Z. J Wang, "Controlling the false discovery rate of the association/causality structure learned with the PC algorithm" The Journal of Machine Learning Research 10 (2009): 475-514.

ddgraph documentation built on Nov. 17, 2017, 10:50 a.m.