rakheon/d2wlasso: Structured variable selection with weighted lasso

Procedures for weighted lasso variable selection where weights help to direct the variable selection procedure. Weights are determined from results of simpler regressions between the outcome variable and each potential predictor variable. Weights implemented in this package are based on q-values, partial correlation coefficients, t-statistics of regression coefficients, and exclusion frequency statistics.

install.packages("remotes")
remotes::install_github("rakheon/d2wlasso")