matie (version 1.2)

matie-package: Measuring Association and Testing Independence Efficiently

Description

A non-parametric measure of association between variables. The association score, $A$, ranges from 0 (when the variables are independent) to 1 (when they are perfectly associated). $A$ is a kind of $R^2$ estimate, and can be thought of as the proportion of variance in one variable explained by another (or explained by a number of other variables - $A$ works for multivariate associations as well).

matie computes $A$ by estimating a generalized $R^2$, which is computed from the ratio of the likelihood of an alternative model (allowing dependence between variables) over the likelihood of a null model (that forces the variables to be independent). See http://arxiv.org/abs/1303.1828 for details.

The variables should be continuous. $A$ is calculated from the ranks of the values, and ties are currently broken randomly. Ordinal data can be handled, but the properties of $A$ have not been thoroughly investigated for non-continuous data with many ties. $A$ is not currently implemented for category-valued data.

Version 1.2 provides underflow protection to fix a bug that appeared at high dimensions in earlier versions.

Details

Package:

matie

Type:

Package

Version:

1.2

Date:

2013-10-11

License:

GPL-3

References

Discovering general multidimensional associations, http://arxiv.org/abs/1303.1828

Examples

    d <- sbd(NULL,n=400,Rsq=0.9)
    aScore <- ma(d)
    pValue <- ma.test(d,aScore)