shrink_eigen: Nonlinear shrinkage of sample eigenvalues

In cernn: Covariance Estimation Regularized by Nuclear Norm Penalties

Description

shrink_eigen shrinks the sample eigenvalues.

Usage

shrink_eigen(d, lambda, alpha, n)

Arguments

d	Vector of sample eigenvalues to shrink. These must be nonnegative.
lambda	Regularization parameter controling amount of shrinkage towards the target.
alpha	Parameter that controls mixture between the trace and inverse trace penalties.
n	The number of observations.

Value

Vector of shrunken eigenvalues.

Examples

set.seed(12345)
nLambda <- 100
lambda <- 10**seq(-2,2,length.out=nLambda)
alpha <- 0.5
n <- 10
p <- 5
d <- sort(2*runif(p))
e <- shrink_eigen(d,lambda,alpha,n)

## Plot regularization paths of eigenvalues
matplot(x=log10(lambda),y=t(e),type='l',ylab='shrunken eigenvalue')

cernn documentation built on May 2, 2019, 6 a.m.