kernel.moment.Rd
\name{kernel.moment}
\alias{kernel.moment}
\title{Moment of Smoothing Kernel}
\description{
Computes the complete or incomplete \eqn{m}th moment of a
smoothing kernel.
}
\usage{
kernel.moment(m, r, kernel = "gaussian")
}
\arguments{
\item{m}{
Exponent (order of moment).
An integer.
}
\item{r}{
Upper limit of integration for the incomplete moment.
A numeric value or numeric vector.
Set \code{r=Inf} to obtain the complete moment.
}
\item{kernel}{
String name of the kernel.
Options are
\code{"gaussian"}, \code{"rectangular"},
\code{"triangular"},
\code{"epanechnikov"},
\code{"biweight"},
\code{"cosine"} and \code{"optcosine"}.
(Partial matching is used).
}
}
\details{
Kernel estimation of a probability density in one dimension
is performed by \code{\link[stats]{density.default}}
using a kernel function selected from the list above.
For more information about these kernels,
see \code{\link[stats]{density.default}}.
The function \code{kernel.moment} computes the partial integral
\deqn{
\int_{-\infty}^r t^m k(t) dt
}{
integral[-Inf][r] t^m k(t) dt
}
where \eqn{k(t)} is the selected kernel, \eqn{r} is the upper limit of
integration, and \eqn{m} is the exponent or order.
Here \eqn{k(t)} is the \bold{standard form} of the kernel,
which has support \eqn{[-1,1]} and
standard deviation \eqn{sigma = 1/c} where \code{c = kernel.factor(kernel)}.
}
\value{
A single number, or a numeric vector of the same length as \code{r}.
}
\seealso{
\code{\link[stats]{density.default}},
\code{\link{dkernel}},
\code{\link{kernel.factor}},
}
\examples{
kernel.moment(1, 0.1, "epa")
curve(kernel.moment(2, x, "epa"), from=-1, to=1)
}
\author{
\adrian
and Martin Hazelton.
}
\keyword{methods}
\keyword{nonparametric}
\keyword{smooth}