cmenet: Bi-Level Selection of Conditional Main Effects
Provides functions for implementing cmenet - a bi-level variable selection method for conditional main effects (see Mak and Wu (2018) <doi:10.1080/01621459.2018.1448828>). CMEs are reparametrized interaction effects which capture the conditional impact of a factor at a fixed level of another factor. Compared to traditional two-factor interactions, CMEs can quantify more interpretable interaction effects in many problems. The current implementation performs variable selection on only binary CMEs; we are working on an extension for the continuous setting.
Documentation:
Downloads:
Linking:
Please use the canonical formhttps://CRAN.R-project.org/package=cmenetto link to this page.