MultipleTesting.AdjustBenjaminiYekutieli Method

public static Vector<double> AdjustBenjaminiYekutieli(
	Vector<double> pValues
)

Public Shared Function AdjustBenjaminiYekutieli ( 
	pValues As Vector(Of Double)
) As Vector(Of Double)

public:
static Vector<double>^ AdjustBenjaminiYekutieli(
	Vector<double>^ pValues
)

static member AdjustBenjaminiYekutieli : 
        pValues : Vector<float> -> Vector<float> 

Parameters

pValues  Vector<Double>

A vector of raw p-values.

Return Value

This method applies the Benjamini-Yekutieli procedure to control the false discovery rate (FDR) under arbitrary dependency structures between tests.

It modifies the Benjamini-Hochberg procedure by incorporating an additional correction factor that accounts for potential dependencies among the hypothesis tests.

The procedure is particularly useful when the independence or positive regression dependency assumptions of the Benjamini-Hochberg procedure cannot be verified. It applies a harmonic number correction factor to ensure FDR control regardless of dependency structure.

This procedure is more conservative than the Benjamini-Hochberg procedure, potentially leading to reduced statistical power. Only use this method when there is evidence of negative or complex dependencies among your tests; otherwise, the Benjamini-Hochberg procedure may provide better power.

Reference