screen_cvtrunc.lasso {nethet}R Documentation

Cross-validated Lasso screening and additional truncation.

Description

Cross-validated Lasso screening and additional truncation.

Usage

screen_cvtrunc.lasso(x, y, k.trunc = 5)

Arguments

x

Predictor matrix.

y

Response vector.

k.trunc

Truncation constant="number of samples per predictor" (default=5).

Details

Computes Lasso coefficients (cross-validation optimal lambda). Truncates smallest coefficients to zero, such that there are no more than n/k.trunc non-zero coefficients.

Value

Active-set.

Author(s)

n.stadler

Examples

screen_cvtrunc.lasso(matrix(rnorm(5000),50,100),rnorm(50))

[Package nethet version 1.16.1 Index]