quadHSIC: Determines the quadratic form of the HSIC unbiased estimator

In kernelPSI: Post-Selection Inference for Nonlinear Variable Selection

Description

For a linear kernel of the outcome L = Y^\top Y, the unbiased HSIC estimator implemented in HSIC can be expressed as a quadratic form of the outcome Y i.e. HSIC(K, L) = Y^\top Q(K) Y. Here, the matrix Q only depends on the kernel similarity matrix K.

Usage

quadHSIC(K)

Arguments

K	kernel similarity matrix

Value

the matrix of the HSIC estimator quadratic form

Examples

n <- 50
p <- 20
X <- matrix(rnorm(n*p), nrow = n, ncol = p)
K <-  X %*% t(X) / p
Q <- quadHSIC(K)

kernelPSI documentation built on Dec. 8, 2019, 1:07 a.m.