devianLM: Detecting Extremal Values in a Normal Linear Model

Provides a method to detect values poorly explained by a Gaussian linear model. The procedure is based on the maximum of the absolute value of the studentized residuals, which is a parameter-free statistic. This approach generalizes several procedures used to detect abnormal values during longitudinal monitoring of biological markers. For methodological details, see: Berthelot G., Saulière G., Dedecker J. (2025). "DEViaN-LM An R Package for Detecting Abnormal Values in the Gaussian Linear Model". HAL Id: hal-05230549. <https://hal.science/hal-05230549>.

Version: 1.0.4
Depends: R (≥ 2.10)
Imports: Rcpp
LinkingTo: Rcpp, RcppArmadillo
Suggests: testthat (≥ 3.0.0)
Published: 2025-10-05
DOI: 10.32614/CRAN.package.devianLM
Author: Guillaume Sauliere ORCID iD [aut, cre], Geoffroy Berthelot [aut], Jérôme Dedecker [aut]
Maintainer: Guillaume Sauliere <guillaumesauliere at hotmail.com>
License: GPL-3
NeedsCompilation: yes
SystemRequirements: OpenMP (optional, for parallel execution)
Materials: NEWS
CRAN checks: devianLM results

Documentation:

Reference manual: devianLM.html , devianLM.pdf

Downloads:

Package source: devianLM_1.0.4.tar.gz
Windows binaries: r-devel: devianLM_1.0.3.zip, r-release: devianLM_1.0.4.zip, r-oldrel: devianLM_1.0.4.zip
macOS binaries: r-release (arm64): devianLM_1.0.3.tgz, r-oldrel (arm64): devianLM_1.0.3.tgz, r-release (x86_64): devianLM_1.0.4.tgz, r-oldrel (x86_64): devianLM_1.0.4.tgz
Old sources: devianLM archive

Linking:

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