corrselect: Correlation-Based and Model-Based Predictor Pruning

Provides functions for predictor pruning using association-based and model-based approaches. Includes corrPrune() for fast correlation-based pruning, modelPrune() for VIF-based regression pruning, and exact graph-theoretic algorithms (Eppstein–Löffler–Strash, Bron–Kerbosch) for exhaustive subset enumeration. Supports linear models, GLMs, and mixed models ('lme4', 'glmmTMB').

Version: 3.0.2
Depends: R (≥ 3.5)
Imports: Rcpp, methods, stats
LinkingTo: Rcpp
Suggests: svglite, GO.db, WGCNA, preprocessCore, impute, energy, minerva, lme4, glmmTMB, MASS, caret, car, carData, microbenchmark, igraph, Boruta, glmnet, corrplot, knitr, rmarkdown, testthat (≥ 3.0.0), tibble
Published: 2025-11-29
DOI: 10.32614/CRAN.package.corrselect
Author: Gilles Colling [aut, cre]
Maintainer: Gilles Colling <gilles.colling051 at gmail.com>
BugReports: https://github.com/gcol33/corrselect/issues
License: MIT + file LICENSE
URL: https://gillescolling.com/corrselect/
NeedsCompilation: yes
Citation: corrselect citation info
Materials: README, NEWS
CRAN checks: corrselect results

Documentation:

Reference manual: corrselect.html , corrselect.pdf
Vignettes: Advanced Topics (source, R code)
Comparison with Alternatives (source, R code)
Quick Start (source, R code)
Theory and Formulation (source, R code)
Complete Workflows: Real-World Examples (source, R code)

Downloads:

Package source: corrselect_3.0.2.tar.gz
Windows binaries: r-devel: corrselect_2.0.1.zip, r-release: corrselect_2.0.1.zip, r-oldrel: corrselect_2.0.1.zip
macOS binaries: r-release (arm64): corrselect_2.0.1.tgz, r-oldrel (arm64): corrselect_2.0.1.tgz, r-release (x86_64): corrselect_3.0.2.tgz, r-oldrel (x86_64): corrselect_3.0.2.tgz
Old sources: corrselect archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=corrselect to link to this page.