Permutation based p-value for the Pearson correlation coefficient {MXM}

Permutation based p-value for the Pearson correlation coefficient

Description

The main task of this test is to provide a p-value PVALUE for the null hypothesis: feature 'X' is independent from 'TARGET' given a conditioning set CS.

Usage

permcor(x1, x2, R = 999) 
permcorrels(y, x, R = 999)

Arguments

x1

A numerical vector.

x2

A numerical vector of the same size as x1.

y

A vector whose length is equal to the number of rows of x.

x

This is a matrix with many variables.

R

The number of permutations to be conducted; set to 999 by default.

Details

This is a computational non parametric (permutation based) correlation coefficient test and is advised to be used when a small sample size is available. If you want to use the Spearman correlation instead, simply provide the ranks of x or of y and x.

Value

For the case of "permcor" a vector consisting of two values, the Pearson correlation and the permutation based p-value. For the "permcorrels" a vector with three values, the Pearson correlation, the test statistic value and the permutation based logged p-value.

Author(s)

Michail Tsagris

R implementation and documentation: Michail Tsagris [e-mail]

References

Legendre Pierre (2000). Comparison of permutation methods for the partial correlation and partial Mantel tests. Journal of Statistical Computation and Simulation 67(1):37-73.

See Also

pc.skel, testIndSpearman, testIndFisher, SES, CondIndTests

Examples

MXM::permcor(iris[, 1], iris[, 2], R = 999)
x <- matrix( rnorm(50 * 100), ncol = 100)
a <- permcorrels(iris[1:50, 1], x)