compInd: Complete Measure of Mutual Multivariate Independence

In steadyICA: ICA and Tests of Independence via Multivariate Distance Covariance

Description

Calculates a complete empirical measure of mutual multivariate independence. Makes use of the utils::combn function.

Usage

compInd(S,group=1:ncol(S),alpha=1)

Arguments

S	The n x d matrix for which you wish to calculate the dependence between d columns from n samples.
group	A length d vector which indicates group membership for each component.
alpha	The index used in calculating the distance between sample observations.

Value

Returns a scalar equal to the empirical multivariate distance between the observed samples, and their grouped counterpart.

Note

Suppose that the each component belongs to exactly one of C groups. This method makes use of the utils::combn and combinat::permn functions. As a result it will be both computationally and memory intensive, even for small to moderate n and small C.

Author(s)

Nicholas James

References

Chasalow, Scott (2012) combinat: Combinatorics Utilities <http://CRAN.R-project.org/package=combinat

See Also

dcovustat, energy::dcov

Examples

library(steadyICA)
library(combinat)
set.seed(100)
S = matrix(rnorm(40),ncol=4)
group = c(1,2,3,3)
compInd(S,group,1)

steadyICA documentation built on May 2, 2019, 7:30 a.m.