checkKKT: Check the KKT condition of lasso solutions

In shuxiaoc/outference: Valid Inference in Linear Regression Corrected for Outlier Removal

Description

This function returns TRUE if and only if the lasso solution satisfy the KKT condition.

Usage

checkKKT(y, X, lambda, beta.hat, tol = 1e-05)

Arguments

y,	the response vector.
X,	the design matrix.
lambda,	the tuning parameter.
beta.hat,	the lasso solution.
tol,	the tolerance.

Details

This function assumes the quadratic loss is scaled by 1/2n. That is,

1/(2n) * ||y-Xβ||_2^2 + λ * ||β||_1

Value

This function returns the desired logical result.

shuxiaoc/outference documentation built on July 8, 2019, 8:30 p.m.