Ordering of Predictors from Coefficient Estimates

This function orders the predictors by decreasing order of the magnitude of the estimated regression coefficient.

Usage

absolute.order(beta)

Arguments

beta

A vector of estimated regression coefficients.

Value

An ordering of the predictors.

Examples

### generate synthetic data
set.seed(1)
n           = 200
p           = 300
X           = matrix(rnorm(n*p),n,p)
beta        = double(p)
beta[1:10]  = 1:10
y           = X %*% beta + rnorm(n)

### glmnet fit
library(glmnet)
#> Loading required package: Matrix
#> Loaded glmnet 4.1-10
beta.lasso = coef(cv.glmnet(X, y))[-1]
lasso.order = absolute.order(beta.lasso)

### ncvreg fit
library(ncvreg)
#> Error in library(ncvreg): there is no package called ‘ncvreg’
beta.scad = c(coef(cv.ncvreg(X, y))[-1])
#> Error in cv.ncvreg(X, y): could not find function "cv.ncvreg"
scad.order = absolute.order(beta.scad)
#> Error: object 'beta.scad' not found