HCPM: Heterogeneous correlation and p-value matrices

In hetcorFS: Unsupervised Feature Selection using the Heterogeneous Correlation Matrix

Heterogeneous correlation and p-value matrices

Description

Extends the traditional correlation matrix (between numerical data) to also include binary and ordinal categorical data and computes the p-values for the tests of uncorrelation.

Usage

HCPM(data = NULL)

Arguments

data	A data frame. Values of type 'numeric' or 'integer' are treated as numerical.

Value

A list with with elements:

cor_mat	An p by p heterogeneous correlation matrix
p_value	An p by p heterogeneous p-values matrix

References

Tortora C., Madhvani S., Punzo A. (2025). Designing unsupervised mixed-type feature selection techniques using the heterogeneous correlation matrix. International Statistical Review. https://doi.org/10.1111/insr.70016

Examples

 
data(ESI)
data=ESI[,-c(1,3,4,6,9)]##removing categorical features
HCPM(data)

hetcorFS documentation built on Nov. 24, 2025, 5:07 p.m.