adjustLi: Li correction of p-values.

In scmamp: Statistical Comparison of Multiple Algorithms in Multiple Problems

Description

This function takes the particular list of possible hypthesis to correct for multiple testing, as defined in Li (2008).

Usage

adjustLi(pvalues)

Arguments

pvalues

Raw p-values in either a vector or a matrix. Note that in case the p-values are in a matrix, all the values are used for the correction. Therefore, if the matrix contains repeated values (as those output by some methods in this package), the repetitions have to be removed.

Details

The test has been implemented according to the version in Garcia et al. (2010), page 2680-2682. This is a simple procedure that provides good results when the highest p-value corrected is below 0.5. However, When the highest p-value is close to 1 the correction is extremely conservative. Actually, when the highest p-value is 1, all the corrected p-values are set at 1. Therefore, it is not advisable to be used under these circumstances.

Value

A vector or matrix with the corrected p-values

References

S. Garcia, A. Fernandez, J. Luengo and F. Herrera (2010) Advanced nonparametric tests for multiple comparison in the design of experiments in computational intelligence and data mining: Experimental analysis of power. Information Sciences, 180, 2044-2064.

J. Li (2008) A two-step rejection prcedure for testing mulitple hypotheses. Journal of Statistical Planning and Inference, 138, 1521-1527.

Examples

data(data_gh_2008)
raw.pvalues <- friedmanPost(data.gh.2008)
adjustLi(raw.pvalues)

scmamp documentation built on May 1, 2019, 10:10 p.m.