Bootstrapped Basis Regression with Oracle Model Selection

Routines for flexible functional form estimation via basis regression, with model selection via the adaptive LASSO or SCAD to prevent overfitting.

polywog is an R package by Brenton Kenkel and Curtis Signorino containing routines for flexible functional form estimation via basis regression with oracle-penalized model selection. You can install the most recent stable version of the package from CRAN by running


in an R console.

To request a feature or report a bug, please visit the package's project page on Github. The package's change log can also be viewed there.

The statistical methods implemented in polywog are documented and applied in the following working papers:


Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.4-1 by Brenton Kenkel, 3 years ago

Browse source code at

Authors: Brenton Kenkel and Curtis S. Signorino

Documentation:   PDF Manual  

GPL (>= 2) license

Imports foreach, Formula, glmnet, iterators, Matrix, ncvreg, Rcpp, stringr

Depends on miscTools

Suggests carData, lattice, rgl

Linking to Rcpp

See at CRAN