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').

Package details

AuthorGilles Colling [aut, cre, cph] (ORCID: <https://orcid.org/0000-0003-3070-6066>)
MaintainerGilles Colling <gilles.colling051@gmail.com>
LicenseMIT + file LICENSE
Version3.2.3
URL https://gillescolling.com/corrselect/
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:
install.packages("corrselect")

Try the corrselect package in your browser

Any scripts or data that you put into this service are public.

corrselect documentation built on July 19, 2026, 1:06 a.m.