The Induced Smoothed Lasso

An implementation of the induced smoothing (IS) idea to lasso regularization models to allow estimation and inference on the model coefficients (currently hypothesis testing only). Linear, logistic, Poisson and gamma regressions with several link functions are implemented. The algorithm is described in the original paper; see and discussed in a tutorial .

Induced Smoothed Lasso

📦 Installation

You can install the development version of islasso from GitHub:

# install.packages("devtools")
devtools::install_github("gianluca-sottile/islasso")

Once installed, load the package:

library(islasso)

🔍 Description

islasso implements the Induced Smoothed Lasso, a robust and interpretable approach for hypothesis testing in high-dimensional linear and generalized linear models.

Key features include:

Efficient Fortran backend for fast computation
Support for Gaussian, Binomial, Poisson, and Gamma families
Smoothed penalization for stable inference
Automatic selection of active variables
Visualization tools powered by ggplot2

🚀 Quick Example

set.seed(123)
sim <- simulXy(n = 100, p = 20, family = "gaussian")
mod <- islasso(y ~ ., data = sim$data)
summary(mod)
plot(mod)

📚 Documentation

📘 Function reference: ?islasso
📄 Vignette: vignette("islasso-intro")
🌐 Website: https://gianluca-sottile.github.io/islasso/

📖 References

Cilluffo G, Sottile G, La Grutta S, Muggeo V (2020). The Induced Smoothed lasso: A practical framework for hypothesis testing in high dimensional regression. Statistical Methods in Medical Research_, 29(3), 765-777. doi:10.1177/0962280219842890