Statistics & Data Science · Regression & Modelling
A topic within Regression & Modelling, itself one of 10 topics in that field and part of Statistics & Data Science.
2 works
Robert Tibshirani
Proposes penalising regression coefficients by their absolute value rather than their square, which shrinks some coefficients to exactly zero and performs variable selection as a side effect of fitting.
link checked 17 Sept 2026Trevor Hastie, Robert Tibshirani & Martin Wainwright
Surveys the lasso and its descendants, the group lasso, the fused lasso, the elastic net, as one family of methods for fitting a model with more predictors than observations.
link checked 17 Sept 2026