English

ensr: R Package for Simultaneous Selection of Elastic Net Tuning Parameters

Computation 2019-07-02 v1

Abstract

Motivation: Elastic net regression is a form of penalized regression that lies between ridge and least absolute shrinkage and selection operator (LASSO) regression. The elastic net penalty is a powerful tool controlling the impact of correlated predictors and the overall complexity of generalized linear regression models. The elastic net penalty has two tuning parameters: λ{\lambda} for the complexity and α{\alpha} for the compromise between LASSO and ridge. The R package glmnet provides efficient tools for fitting elastic net models and selecting λ{\lambda} for a given α.{\alpha}. However, glmnet does not simultaneously search the λα{\lambda} - {\alpha} space for the optional elastic net model. Results: We built the R package ensr, elastic net searcher. enser extends the functionality of glment to search the λα{\lambda} - {\alpha} space and identify an optimal λα{\lambda} - {\alpha} pair. Availability: ensr is available from the Comprehensive R Archive Network at https://cran.r-project.org/package=ensr

Keywords

Cite

@article{arxiv.1907.00914,
  title  = {ensr: R Package for Simultaneous Selection of Elastic Net Tuning Parameters},
  author = {Peter E. DeWitt and Tellen D. Bennett},
  journal= {arXiv preprint arXiv:1907.00914},
  year   = {2019}
}