A Comprehensive R Package for Best Subset Selection
bestridge: A Comprehensive R Package for Best Subset Selection
Best subset ridge regression
Provides estimated coefficients from a fitted "bsrr" object.
Extract the deviance from a "bsrr.one" object.
Generate simulated data
breast cancer data set
Extract the log-likelihood from a "bsrr.one" object.
Lymphoma patients data set
Produces a coefficient profile plot of the coefficient or loss functio...
make predictions from a "bsrr" object.
print method for a "bsrr" object
Factors associated with prostate specific antigen
Risk factors associated with heart disease
summary method for a "bsrr" object
The bestridge package is designed to provide a one-stand service for users to successfully carry out best ridge regression in various complex situations via the primal dual active set algorithm proposed by Wen, C., Zhang, A., Quan, S. and Wang, X. (2020) <doi:10.18637/jss.v094.i04>. This package allows users to perform the regression, classification, count regression and censored regression for (ultra) high dimensional data, and it also supports advanced usages like group variable selection and nuisance variable selection.