grpregOverlap: Penalized Regression Models with Overlapping Grouped Covariates
Fit the regularization path of linear, logistic or Cox models with
overlapping grouped covariates based on the latent group lasso approach. Latent
group MCP/SCAD as well as bi-level selection methods, namely the group exponential
lasso and the composite MCP are also available. This package serves as an
extension of R package 'grpreg' (by Dr. Patrick Breheny <patrick-breheny@uiowa.edu>)
for grouped variable selection involving overlaps between groups.
Downloads:
Reverse dependencies:
Linking:
Please use the canonical form
https://CRAN.R-project.org/package=grpregOverlap
to link to this page.