LassoBacktracking (version 1.1)
Modelling Interactions in High-Dimensional Data with
Backtracking
Description
Implementation of the algorithm introduced in Shah, R. D.
(2016) .
Data with thousands of predictors can be handled. The algorithm
performs sequential Lasso fits on design matrices containing
increasing sets of candidate interactions. Previous fits are used to greatly
speed up subsequent fits, so the algorithm is very efficient.