oscar: Optimal Subset Cardinality Regression (OSCAR) Models Using the L0-Pseudonorm

Optimal Subset Cardinality Regression (OSCAR) models offer regularized linear regression using the L0-pseudonorm, conventionally known as the number of non-zero coefficients. The package estimates an optimal subset of features using the L0-penalization via cross-validation, bootstrapping and visual diagnostics. Effective Fortran implementations are offered along the package for finding optima for the DC-decomposition, which is used for transforming the discrete L0-regularized optimization problem into a continuous non-convex optimization task. These optimization modules include DBDC ('Double Bundle method for nonsmooth DC optimization' as described in Joki et al. (2018) <doi:10.1137/16M1115733>) and LMBM ('Limited Memory Bundle Method for large-scale nonsmooth optimization' as in Haarala et al. (2004) <doi:10.1080/10556780410001689225>). Multiple regression model families are supported: Cox, logistic, and Gaussian.

Version: 1.0.4
Depends: R (≥ 3.6.0)
Imports: graphics, grDevices, hamlet, Matrix, methods, stats, survival, utils
Suggests: ePCR, glmnet, knitr, pROC, rmarkdown
Published: 2022-05-23
Author: Teemu Daniel Laajala ORCID iD [aut, cre], Kaisa Joki [aut], Anni Halkola [aut]
Maintainer: Teemu Daniel Laajala <teelaa at utu.fi>
BugReports: https://github.com/Syksy/oscar/issues
License: GPL-3
URL: https://github.com/Syksy/oscar
NeedsCompilation: yes
Materials: README
CRAN checks: oscar results


Reference manual: oscar.pdf
Vignettes: Example use of the OSCAR package


Package source: oscar_1.0.4.tar.gz
Windows binaries: r-devel: oscar_1.0.4.zip, r-release: oscar_1.0.4.zip, r-oldrel: oscar_1.0.4.zip
macOS binaries: r-release (arm64): oscar_1.0.4.tgz, r-oldrel (arm64): oscar_1.0.4.tgz, r-release (x86_64): oscar_1.0.4.tgz, r-oldrel (x86_64): oscar_1.0.4.tgz
Old sources: oscar archive


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