fda.usc (version 2.0.1)

Kernel: Symmetric Smoothing Kernels.

Description

Represent symmetric smoothing kernels:: normal, cosine, triweight, quartic and uniform.

Usage

Kernel(u, type.Ker = "Ker.norm")

Arguments

u

Data.

type.Ker

Type of Kernel. By default normal kernel.

Value

Returns symmetric kernel.

Details

Ker.norm=dnorm(u)
Ker.cos=ifelse(abs(u)<=1,pi/4*(cos(pi*u/2)),0)
Ker.epa=ifelse(abs(u)<=1,3/4*(1-u^2),0)
Ker.tri=ifelse(abs(u)<=1,35/32*(1-u^2)^3,0)
Ker.quar=ifelse(abs(u)<=1,15/16*(1-u^2)^2,0)
Ker.unif=ifelse(abs(u)<=1,1/2,0)

Type of kernel:

Normal Kernel: Ker.norm
Cosine Kernel: Ker.cos
Epanechnikov Kernel: Ker.epa
Triweight Kernel: Ker.tri
Quartic Kernel: Ker.quar
Uniform Kernel: Ker.unif

References

Ferraty, F. and Vieu, P. (2006). Nonparametric functional data analysis. Springer Series in Statistics, New York.

Hardle, W. Applied Nonparametric Regression. Cambridge University Press, 1994.

Examples

Run this code
# NOT RUN {
y=qnorm(seq(.1,.9,len=100))
a<-Kernel(u=y)
b<-Kernel(type.Ker="Ker.tri",u=y)
c=Ker.cos(y)
# }

Run the code above in your browser using DataCamp Workspace