Q_func: This Q function is introduced in the original FISTA paper. It...

In kevinbai92/LSvarEstimate: Low Rank Plus Sparse VAR Model Estimation

Description

This Q function is introduced in the original FISTA paper. It is the quadratic approximation of the objective function.

Usage

Q_func(x, y, A, b, L, AtA, Atb)

Arguments

x	coefficient vector (matrix)
y	an alternative coefficient vector (proximal)
A	design matrix in linear regression
b	target vector in linear regression
L	learning step size
AtA	crossproduct of A
Atb	crossproduct of A and b

Details

Q function contributes the quadratic approximation to the objective function (loss + penalties)

Value

the value of Q function

Examples

## Not run: 
set.seed(1)
x <- stats::rnorm(10)
y <- stats::rnorm(10)
A <- matrix(stats::rnorm(200), nrow = 20, ncol = 10)
b <- stats:rnorm(20)
L <- 0.2
AtA <- t(A) %*% A
Atb <- t(A) %*% b
Q_func(x, y, A, b, L, AtA, Atb)

## End(Not run)

kevinbai92/LSvarEstimate documentation built on May 8, 2020, 1:04 a.m.