lassogrp-package: Lasso regression including group lasso and adaptive lasso

Description Details Author(s) References See Also Examples

Description

lassogrp implements the Lasso regression method for several regression types (ordinary, logistic, Poisson and user-provided models). It offers the group lasso which is suitable when factors appear in the model. It also provides a user-friendly way to perform the adaptive lasso, which consists of two calls to lasso, the second one using a weighted L1 penalization term, where the weights depend on the results of the first call.

Details

The DESCRIPTION file: This package was not yet installed at build time.
Index: This package was not yet installed at build time.

Aim: Replaces the grplasso package and extends its functionality.

The lassogrp package includes methods for printing and plotting as well as a cross validation function and functions to extract specified fits from the lasso results.

Author(s)

Werner A. Stahel, ETH Zurich, stahel@stat.math.ethz.ch, based on code of Lukas Meier and cosmetics by Martin Maechler.

References

Lukas Meier, Sara van de Geer and Peter B\"uhlmann (2008), The Group Lasso for Logistic Regression, Journal of the Royal Statistical Society, 70 (1), 53–71

See Also

The packages lars, glmnet.

Examples

1
2
data(splice)
fit.splice <- lasso(y ~ ., data = splice, model = 'binomial')

lassogrp documentation built on July 25, 2021, 3 a.m.