epspath: Fit the entire 'epsilon' path for Support Vector Regression

In svrpath: The SVR Path Algorithm

Description

The Suport Vector Regression (SVR) employs epsilon-intensive loss which ignores errors smaller than epsilon. This algorithm computes the entire paths for SVR solution as a function of epsilon at a given regularization parameter lambda, which we call epsilon path.

Usage

epspath(x, y, lambda = 1, kernel.function = radial.kernel,
  param.kernel = 1, ridge = 1e-08, eps = 1e-07, eps.min = 1e-08, ...)

Arguments

x	The data matrix (n x p) with n rows (observations) on p variables (columns)
y	The real number valued response variable
lambda	The regularization parameter value.
kernel.function	User defined kernel function. See svmpath.
param.kernel	Parameter(s) of the kernels. See svmpath.
ridge	Sometimes the algorithm encounters singularities; in this case a small value of ridge can help, default is ridge = 1e-8
eps	A small machine number which is used to identify minimal step sizes
eps.min	The smallest value of epsilon for termination of the algorithm. Default is eps.min = 1e-8
...	Generic compatibility

Value

An 'epspath' object is returned.

Author(s)

Do Hyun Kim, Seung Jun Shin

See Also

predict.epspath, plot.epspath, svrpath

Examples

set.seed(1)
n <- 30
p <- 50

x <- matrix(rnorm(n*p), n, p)
e <- rnorm(n, 0, 1)
beta <- c(1, 1, rep(0, p-2))
y <- x %*% beta + e
lambda <- 1
eobj <- epspath(x, y, lambda = lambda)

svrpath documentation built on May 2, 2019, 9:13 a.m.