compInd {steadyICA} | R Documentation |
Complete Measure of Mutual Multivariate Independence
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
Examples
library(steadyICA)
library(combinat)
set.seed(100)
S = matrix(rnorm(40),ncol=4)
group = c(1,2,3,3)
compInd(S,group,1)