active_set: Determine the active set variables via lasso regression but...

In Marga8/HDGCvar: Granger Causality Testing in High Dimensional Vector Autoregressive Models

Determine the active set variables via lasso regression but it augments also the lasso selection.

Description

Determine the active set variables via lasso regression but it augments also the lasso selection.

Usage

active_set(i, d, p, X_index, y, z, z_a = NULL, bound = 0.5 * NROW(z))

Arguments

i	determines which column of y is considered
d	the decided lag-augmentation of the interest variables
p	the lag length
X_index	index of the Granger causing variables
y	vector/matrix of dependent variable
z	vector/matrix of regressors
z_a	those variables that should not be penalized
bound	upper bound on tuning parameter lambda

Value

the active set estimated via lasso using BIC as penalty

Marga8/HDGCvar documentation built on May 25, 2024, 11:12 a.m.