In ryan-thompson/grpsel: Group Subset Selection

knitr::opts_chunk$set(echo = T)

grpsel

Overview

An R package for sparse regression modelling with grouped predictors (including overlapping groups). grpsel uses the group subset selection penalty, usually leading to excellent selection and prediction. Optionally, the group subset penalty can be combined with a group lasso or ridge penalty for added shrinkage. Linear and logistic regression are currently supported. See this paper for more information.

Installation

To install the latest stable version from CRAN, run the following code: ``` {r, eval = F} install.packages('grpsel')

To install the latest development version from GitHub, run the following code:
``` {r, eval = F}
devtools::install_github('ryan-thompson/grpsel')

Usage

The grpsel() function fits a group subset regression model for a sequence of tuning parameters. The cv.grpsel() function provides a convenient way to automatically cross-validate these parameters.

library(grpsel)

# Generate some grouped data
set.seed(123)
n <- 100 # Number of observations
p <- 10 # Number of predictors
g <- 5 # Number of groups
group <- rep(1:g, each = p / g) # Group structure
beta <- numeric(p)
beta[which(group %in% 1:2)] <- 1 # First two groups are nonzero
x <- matrix(rnorm(n * p), n, p)
y <- x %*% beta + rnorm(n)

# Fit the group subset selection regularisation path
fit <- grpsel(x, y, group)
coef(fit, lambda = 0.05)

# Cross-validate the group subset selection regularisation path
fit <- cv.grpsel(x, y, group)
coef(fit)