Implementation of the algorithm introduced in "Shah, R. D. (2016) Modelling interactions in high-dimensional data with Backtracking, JMLR, to appear". Data with thousands of predictors can be handled. The algorithm performs sequential Lasso (Tibshirani, 1996) 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.
|Author||Rajen Shah [aut, cre]|
|Date of publication||2016-04-14 14:49:49|
|Maintainer||Rajen Shah <email@example.com>|
|License||GPL (>= 2)|