Revision d606122dc24b56ecf537d55eda38f4bf5ac4de1f authored by Adrian Baddeley on 25 October 2010, 10:40:51 UTC, committed by cran-robot on 25 October 2010, 10:40:51 UTC
1 parent 66bc933
reach.Rd
\name{reach}
\alias{reach}
\alias{reach.ppm}
\alias{reach.interact}
\alias{reach.rmhmodel}
\title{Interaction Distance of a Point Process}
\description{
Computes the interaction distance of a point process.
}
\usage{
reach(x, \dots)
\method{reach}{ppm}(x, \dots, epsilon=0)
\method{reach}{interact}(x, \dots)
\method{reach}{rmhmodel}(x, \dots)
}
\arguments{
\item{x}{Either a fitted point process model (object of class
\code{"ppm"}), an interpoint interaction (object of class
\code{"interact"}) or a point process model for simulation
(object of class \code{"rmhmodel"}).
}
\item{epsilon}{
Numerical threshold below which interaction is treated as zero.
See details.
}
\item{\dots}{
Other arguments are ignored.
}
}
\value{
The interaction distance, or \code{NA} if this cannot be
computed from the information given.
}
\details{
The `interaction distance' or `interaction range' of a point process
model is the smallest distance \eqn{D} such that any two points in the
process which are separated by a distance greater than \eqn{D} do not
interact with each other.
For example, the interaction range of a Strauss process
(see \code{\link{Strauss}})
with parameters \eqn{\beta,\gamma,r}{beta,gamma,r} is equal to
\eqn{r}, unless \eqn{\gamma=1}{gamma=1} in which case the model is
Poisson and the interaction
range is \eqn{0}.
The interaction range of a Poisson process is zero.
The interaction range of the Ord threshold process
(see \code{\link{OrdThresh}}) is infinite, since two points \emph{may}
interact at any distance apart.
The function \code{reach(x)} is generic, with methods
for the case where \code{x} is
\itemize{
\item
a fitted point process model
(object of class \code{"ppm"}, usually obtained from the model-fitting
function \code{\link{ppm}});
\item
an interpoint interaction structure (object of class
\code{"interact"}), created by one of the functions
\code{\link{Poisson}},
\code{\link{Strauss}},
\code{\link{StraussHard}},
\code{\link{MultiStrauss}},
\code{\link{MultiStraussHard}},
\code{\link{Softcore}},
\code{\link{DiggleGratton}},
\code{\link{Pairwise}},
\code{\link{PairPiece}},
\code{\link{Geyer}},
\code{\link{LennardJones}},
\code{\link{Saturated}},
\code{\link{OrdThresh}}
or
\code{\link{Ord}};
\item
a point process model for simulation (object of class
\code{"rmhmodel"}), usually obtained from \code{\link{rmhmodel}}.
}
When \code{x} is an \code{"interact"} object,
\code{reach(x)} returns the maximum possible interaction range
for any point process model with interaction structure given by \code{x}.
For example, \code{reach(Strauss(0.2))} returns \code{0.2}.
When \code{x} is a \code{"ppm"} object,
\code{reach(x)} returns the interaction range
for the point process model represented by \code{x}.
For example, a fitted Strauss process model
with parameters \code{beta,gamma,r} will return
either \code{0} or \code{r}, depending on whether the fitted
interaction parameter \code{gamma} is equal or not equal to 1.
For some point process models, such as the soft core process
(see \code{\link{Softcore}}), the interaction distance is
infinite, because the interaction terms are positive for all
pairs of points. A practical solution is to compute
the distance at which the interaction contribution
from a pair of points falls below a threshold \code{epsilon},
on the scale of the log conditional intensity. This is done
by setting the argument \code{epsilon} to a positive value.
}
\seealso{
\code{\link{ppm}},
\code{\link{Poisson}},
\code{\link{Strauss}},
\code{\link{StraussHard}},
\code{\link{MultiStrauss}},
\code{\link{MultiStraussHard}},
\code{\link{Softcore}},
\code{\link{DiggleGratton}},
\code{\link{Pairwise}},
\code{\link{PairPiece}},
\code{\link{Geyer}},
\code{\link{LennardJones}},
\code{\link{Saturated}},
\code{\link{OrdThresh}},
\code{\link{Ord}},
\code{\link{rmhmodel}}
}
\examples{
reach(Poisson())
# returns 0
reach(Strauss(r=7))
# returns 7
data(swedishpines)
fit <- ppm(swedishpines, ~1, Strauss(r=7))
reach(fit)
# returns 7
reach(OrdThresh(42))
# returns Inf
reach(MultiStrauss(1:2, matrix(c(1,3,3,1),2,2)))
# returns 3
}
\author{Adrian Baddeley
\email{adrian@maths.uwa.edu.au}
\url{http://www.maths.uwa.edu.au/~adrian/}
and Rolf Turner
\email{r.turner@auckland.ac.nz}
}
\keyword{spatial}
\keyword{models}
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...