ASE_reg: The ASE function for the local linear estimator (LLE) in the...

In OSCV: One-Sided Cross-Validation

Description

Computing ASE(h), the value of the ASE function for the local linear estimator in the regression context, for the given vector of h values.

Usage

ASE_reg(h, desx, y, rx)

Arguments

h	numerical vector of bandwidth values,
desx	numerical vecror of design points,
y	numerical vecror of data points corresponding to the design points desx,
rx	numerical vecror of values of the regression function at desx.

Details

The average squared error (ASE) is used as a measure of performace of the local linear estimator based on the Gaussian kernel.

Value

The vector of values of ASE(h) for the correponsing vector of h values.

References

Hart, J.D. and Yi, S. (1998) One-sided cross-validation. Journal of the American Statistical Association, 93(442), 620-631.

See Also

loclin, h_ASE_reg, CV_reg, OSCV_reg.

Examples

## Not run: 
# Example (ASE function for a random sample of size n=100 generated from the function reg3 that
# has six cusps. The function originates from the article of Savchuk et al. (2013).
# The level of the added Gaussian noise is sigma=1/1000).
n=100
dx=(1:n-0.5)/n
regx=reg3(dx)
ydat=regx+rnorm(n,sd=1/1000)
harray=seq(0.003,0.05,len=300)
ASEarray=ASE_reg(harray,dx,ydat,regx)
hmin=round(h_ASE_reg(dx,ydat,regx),digits=4)
dev.new()
plot(harray,ASEarray,'l',lwd=3,xlab="h",ylab="ASE",main="ASE function for a random sample
from r3",cex.lab=1.7,cex.axis=1.7,cex.main=1.5)
legend(0.029,0.0000008,legend=c("n=100","sigma=1/1000"),cex=1.7,bty="n")
legend(0.005,0.000002,legend=paste("h_ASE=",hmin),cex=2,bty="n")

## End(Not run)

Example output

Loading required package: mc2d
Loading required package: mvtnorm

Attaching package: 'mc2d'

The following objects are masked from 'package:base':

    pmax, pmin

OSCV documentation built on May 2, 2019, 6:44 a.m.