lin.int.ineq: Initialize Linear Programming Fit with Inequality Constraints

In PACLasso: Penalized and Constrained Lasso Optimization

Description

This function is called internally by lars.ineq to get the linear programming initial fit if the user requests implementation of the algorithm starting at the largest lambda value and proceeding backwards.

Usage

lin.int.ineq(C.full, b)

Arguments

C.full	complete constraint matrix C (with inequality constraints of the form C.full*beta >= b))
b	constraint vector b

Value

beta the initial beta vector of coefficients to use for the PaC algorithm

Examples

random_data = generate.data(n = 500, p = 20, m = 10)
lin_start = lin.int.ineq(random_data$C.full, random_data$b)
lin_start

PACLasso documentation built on May 2, 2019, 2:29 p.m.