Vtr: Vtr

In zhaolotelli/SC19017: SC19017 final report package for Statistical computing

Description

Compute the e.d.r. direction on the training data

Usage

Vtr(Xtrain, Xsple = Xtrain, p, ei, kernel = "AG", sigma = 2)

Arguments

Xtrain	training data
Xsple	if data is too large, can choose some sample of training data
p	dimensions of e.d.r. directions
ei	eigen analysis result of Ei
kernel	kernel functions, "AG" means AGKernel, "G" means GKernel, "P" means PKernel
sigma	bandwidth

Value

a matrix with column vector being e.d.r. direction on the training set

Vtr	column vectors are the e.d.r. directions of training data

Examples

## Not run: 
X <- matrix(rnorm(30), 15, 2)
Kx <- KX(X)
y <- sample(1:15, 15)
J <- Jy(y, h = 5)
ei <- Ei(Kx, J, s = 0.1)
Vtr(X, X, p = 2, ei)

## End(Not run)

zhaolotelli/SC19017 documentation built on Jan. 3, 2020, 9:19 p.m.