FindCorr: Determine Highly Correlated Variables

In AndriSignorell/DescTools: Tools for Descriptive Statistics

Determine Highly Correlated Variables

Description

This function searches through a correlation matrix and returns a vector of integers corresponding to columns to remove to reduce pair-wise correlations.

Usage

FindCorr(x, cutoff = .90, verbose = FALSE)

Arguments

x	A correlation matrix
cutoff	A numeric value for the pair-wise absolute correlation cutoff
verbose	A boolean for printing the details

Details

The absolute values of pair-wise correlations are considered. If two variables have a high correlation, the function looks at the mean absolute correlation of each variable and removes the variable with the largest mean absolute correlation.

There are several function in the subselect package that can also be used to accomplish the same goal. However the package was removed from CRAN and available in the archives.

Value

A vector of indices denoting the columns to remove. If no correlations meet the criteria, numeric(0) is returned.

Author(s)

Original R code by Dong Li, modified by Max Kuhn

References

Max Kuhn. Contributions from Jed Wing, Steve Weston, Andre Williams, Chris Keefer, Allan Engelhardt, Tony Cooper, Zachary Mayer and the R Core Team (2014). caret: Classification and Regression Training. R package version 6.0-35. https://cran.r-project.org/package=caret

Examples

corrMatrix <- diag(rep(1, 5))
corrMatrix[2, 3] <- corrMatrix[3, 2] <- .7
corrMatrix[5, 3] <- corrMatrix[3, 5] <- -.7
corrMatrix[4, 1] <- corrMatrix[1, 4] <- -.67

corrDF <- expand.grid(row = 1:5, col = 1:5)
corrDF$correlation <- as.vector(corrMatrix)
PlotCorr(xtabs(correlation ~ ., corrDF), las=1, border="grey")

FindCorr(corrMatrix, cutoff = .65, verbose = TRUE)

FindCorr(corrMatrix, cutoff = .99, verbose = TRUE)

# d.pizza example
m <- cor(data.frame(lapply(d.pizza, as.numeric)), use="pairwise.complete.obs")
FindCorr(m, verbose = TRUE)
m[, FindCorr(m)]

AndriSignorell/DescTools documentation built on Nov. 11, 2024, 12:11 p.m.