https://github.com/cran/RandomFields
Tip revision: 683e381531c37e8e7224edd899422f119d926418 authored by Martin Schlather on 21 January 2014, 00:00:00 UTC
version 3.0.10
version 3.0.10
Tip revision: 683e381
RMmastein.Rd
\name{RMmastein}
\alias{RMmastein}
\title{Ma-Stein operator}
\description{
\command{\link{RMmastein}} is a univariate stationary covariance model
depending on a variogram or covariance model on the real axis.
The corresponding covariance function only depends on the difference
\eqn{h}{h} between two points and is given by
\deqn{C(h, t)=\frac{\Gamma(\nu + \phi(t))\Gamma(\nu + \delta)}{
\Gamma(\nu + \phi(t) + \delta) \Gamma(\nu)} W_{\nu +
\phi(t)}(\|h -Vt\|)}{C(h, t)= [ Gamma(nu + phi(t))Gamma(nu +
delta) ] / [Gamma(nu + phi(t) + delta) Gamma(nu) ] W_{nu +
phi(t)}(|h - Vt|)}
if \eqn{\phi} is a variogram model.
It is given by
\deqn{C(h, t)=\frac{\Gamma(\nu + \phi(0)-\phi(t))\Gamma(\nu + \delta)}{
\Gamma(\nu + \phi(0)-\phi(t) + \delta) \Gamma(\nu)} W_{\nu +
\phi(t)}(\|h -Vt\|)}{C(h, t)= [ Gamma(nu + phi(0)-phi(t))Gamma(nu +
delta) ] / [Gamma(nu + phi(0)-phi(t) + delta) Gamma(nu) ] W_{nu + phi(0)-phi(t)}(|h - Vt|)}
if \eqn{\phi} is a covariance model.
Here \eqn{\Gamma} is the Gamma function; \eqn{W} is the Whittle-Matern
model (RMwhittle).
}
\usage{
RMmastein(phi, nu, delta, var, scale, Aniso, proj)
}
\arguments{
\item{phi}{an \command{\link{RMmodel}} on the real axis}
\item{nu}{numerical value; positive; smoothness parameter of the
Whittle-Matern model (for \eqn{t=0})}
\item{delta}{a numerical value; \eqn{\delta} must be greater than or
equal to half the dimension of \eqn{h}}
\item{var,scale,Aniso,proj}{optional parameters; same meaning for any
\command{\link{RMmodel}}. If not passed, the above
covariance function remains unmodified.}
}
\details{
See Stein (2005) formula (12).
Instead of the velocity parameter \eqn{V} in the original model
description, a preceeding anisotropy matrix is chosen appropriately:
\deqn{\left( \begin{array}{cc} A & -V \\ 0 &
1\end{array}\right)}{matrix(c(A, -V, 0, 1), nr=2, by=TRUE)}
A is a spatial transformation matrix.
(I.e. (x,t) is multiplied from left on the above matrix and
the first elements of the obtained vector are intepreted as
new spatial components and only these components are used to form
the argument in the Whittle-Matern function.)
The last component in the new coordinates is the time which is
passed to \eqn{\phi}{phi}. (Velocity is assumed to be zero in
the new coordinates.)
Note, that for numerical reasons, \eqn{\nu+\phi+d} may not exceed
the value 80.0. If exceeded the algorithm fails.
}
\value{
\command{\link{RMmastein}} returns an object of class \code{\link[=RMmodel-class]{RMmodel}}
}
\references{
\itemize{
\item Ma, C. (2003)
Spatio-temporal covariance functions generated by mixtures.
\emph{Math. Geol.}, \bold{34}, 965-975.
\item Stein, M.L. (2005) Space-time covariance functions. \emph{JASA},
\bold{100}, 310-321.
}
}
\author{Martin Schlather, \email{schlather@math.uni-mannheim.de}
}
\seealso{
\command{\link{RMwhittle}},
\command{\link{RMmodel}},
\command{\link{RFsimulate}},
\command{\link{RFfit}}.
}
\keyword{spatial}
\keyword{models}
\examples{
RFoptions(seed=0)
model <- RMmastein(RMgauss(), nu=1, delta=10)
x <- seq(0, 10, if (interactive()) 0.1 else 3)
plot(RFsimulate(model, x=x, y=x))
\dontshow{RFoptions(seed=NA)}
}