cvLM: Cross-Validation for Linear and Ridge Regression Models

Implements cross-validation methods for linear and ridge regression models. The package provides grid-based selection of the ridge penalty parameter using Singular Value Decomposition (SVD) and supports K-fold cross-validation, Leave-One-Out Cross-Validation (LOOCV), and Generalized Cross-Validation (GCV). Computations are implemented in C++ via 'RcppArmadillo' with optional parallelization using 'RcppParallel'. The methods are suitable for high-dimensional settings where the number of predictors exceeds the number of observations.

Version:	2.0.0
Imports:	stats, Rcpp (≥ 1.0.13), RcppParallel (≥ 5.1.8)
LinkingTo:	Rcpp, RcppArmadillo, RcppParallel
Suggests:	boot, RhpcBLASctl, testthat (≥ 3.0.0)
Published:	2026-02-03
DOI:	10.32614/CRAN.package.cvLM
Author:	Philip Nye [aut, cre]
Maintainer:	Philip Nye <phipnye at proton.me>
License:	MIT + file LICENSE
URL:	https://github.com/phipnye/CV-LM
NeedsCompilation:	yes
SystemRequirements:	GNU make, C++17
Materials:	README, NEWS
CRAN checks:	cvLM results

Documentation:

Reference manual:

cvLM.html , cvLM.pdf

Downloads:

Package source:	cvLM_2.0.0.tar.gz
Windows binaries:	r-devel: cvLM_1.0.4.zip, r-release: cvLM_1.0.4.zip, r-oldrel: cvLM_1.0.4.zip
macOS binaries:	r-release (arm64): cvLM_2.0.0.tgz, r-oldrel (arm64): cvLM_2.0.0.tgz, r-release (x86_64): cvLM_2.0.0.tgz, r-oldrel (x86_64): cvLM_2.0.0.tgz
Old sources:	cvLM archive

Linking:

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