getCandidates: Get candidates from marginal ordering of variables

In JieGroup/bc: Model Selection with Bridge Criterion

Description

This function uses a weighting method to assign weights to each subset in a list of candidate subsets, and then obtain marginal weights for variables to obtain a new candidate set with nested subsets.

Usage

getCandidates(initCandidates, dat, weight_method = "ARM")

Arguments

initCandidates	List of vectors each indicating an initial subset
dat	List containing design Matrix X and observation Vector y
weight_method	String indicating which method to use for weighting

Value

candidates List of vectors each representing a new candidate subset

w_var Vector of each variable's marginal importance

ind_var Vector of each variable's index in the original system 1...p

JieGroup/bc documentation built on June 1, 2019, 12:48 p.m.