moreparty (version 0.4)

BivariateAssoc: Bivariate association measures for supervised learning tasks.

Description

Computes bivariate association measures between a response and predictor variables (and, optionnaly, between every pairs of predictor variables.)

Usage

BivariateAssoc(Y, X, xx = TRUE)

Value

A list of the following items :

YX

: a table with the association measures between the response and predictor variables

XX

: a table with the association measures between every couples of predictor variables

In each table :

measure

: name of the "standard" association measure

assoc

: value of the "standard" association measure

p.value

: p-value from the permutation test

criterion

: p-value from the permutation test transformed as -log(1-p), which serves to sort rows

Arguments

Y

the response variable

X

the predictor variables

xx

whether the association measures should be computed for couples of predictor variables (default) or not. With a lot of predictors, consider setting xx to FALSE (for reasons of computation time).

Author

Nicolas Robette

Details

For each pair of variable, a permutation test is computed, following the framework used in conditional inference trees to choose a splitting variable. This test produces a p-value, transformed as -log(1-p) for reasons of comparison stability. The function also computes a "standard" association measure : kenddal's tau correlation for pairs of numeric variables, Cramer's V for pairs of factors and eta-squared for pairs numeric-factor.

References

Hothorn T, Hornik K, Van De Wiel MA, Zeileis A. "A lego system for conditional inference". The American Statistician. 60:257–263, 2006.

Hothorn T, Hornik K, Zeileis A. "Unbiased Recursive Partitioning: A Conditional Inference Framework". Journal of Computational and Graphical Statistics, 15(3):651-674, 2006.

See Also

ctree

Examples

  data(iris)
  iris2 = iris
  iris2$Species = factor(iris$Species == "versicolor")
  BivariateAssoc(iris2$Species,iris2[,1:4])