Revision 94dac762f5b1607c843aa97ea6a6caa68343fdb1 authored by Adrian Baddeley on 22 March 2019, 12:40:03 UTC, committed by cran-robot on 22 March 2019, 12:40:03 UTC
1 parent 80e2bf6
fryplot.Rd
\name{fryplot}
\alias{fryplot}
\alias{frypoints}
\title{Fry Plot of Point Pattern}
\description{
Displays the Fry plot (Patterson plot) of a spatial point pattern.
}
\usage{
fryplot(X, ..., width=NULL, from=NULL, to=NULL, axes=FALSE)
frypoints(X, from=NULL, to=NULL, dmax=Inf)
}
\arguments{
\item{X}{
A point pattern (object of class \code{"ppp"}) or something
acceptable to \code{\link{as.ppp}}.
}
\item{\dots}{Optional arguments to control the appearance of the
plot.}
\item{width}{Optional parameter indicating the width of a box
for a zoomed-in view of the Fry plot near the origin.}
\item{from,to}{
Optional. Subset indices specifying which points of \code{X}
will be considered when forming the vectors (drawn from each point of
\code{from}, to each point of \code{to}.)
}
\item{axes}{
Logical value indicating whether to draw axes,
crossing at the origin.
}
\item{dmax}{
Maximum distance between points. Pairs at greater distances
do not contribute to the result. The default means there is
no maximum distance.
}
}
\details{
The function \code{fryplot} generates a Fry plot (or Patterson plot);
\code{frypoints} returns the points of the Fry plot as a point pattern
dataset.
Fry (1979) and Hanna and Fry (1979) introduced a manual graphical method for
investigating features of a spatial point pattern of mineral deposits.
A transparent sheet, marked
with an origin or centre point, is placed over the point pattern.
The transparent sheet is shifted so that the origin lies over one of
the data points, and the positions of all the \emph{other} data points
are copied onto the transparent sheet. This procedure is repeated for
each data point in turn. The resulting plot (the Fry plot)
is a pattern of \eqn{n(n-1)} points, where \eqn{n} is the original number
of data points. This procedure was previously proposed by
Patterson (1934, 1935) for studying inter-atomic distances in
crystals, and is also known as a Patterson plot.
The function \code{fryplot} generates the Fry/Patterson plot.
Standard graphical parameters
such as \code{main}, \code{pch},
\code{lwd}, \code{col}, \code{bg}, \code{cex} can be used to control
the appearance of the plot.
To zoom in (to view only a subset of the Fry plot at higher
magnification), use the argument \code{width} to specify the width
of a rectangular field of view centred at the origin, or the standard
graphical arguments \code{xlim} and \code{ylim} to specify another
rectangular field of view. (The actual field of view may be slightly
larger, depending on the graphics device.)
The function \code{frypoints} returns the points of the Fry
plot as a point pattern object. There may be a large number of points
in this pattern, so this function should be used only if further
analysis of the Fry plot is required.
Fry plots are particularly useful for recognising anisotropy in
regular point patterns. A void around the origin in the Fry plot
suggests regularity (inhibition between points) and the shape of the
void gives a clue to anisotropy in the pattern. Fry plots are also
useful for detecting periodicity or rounding of the spatial
coordinates.
In mathematical terms, the Fry plot of a point pattern \code{X}
is simply a plot of the vectors \code{X[i] - X[j]} connecting all
pairs of distinct points in \code{X}.
The Fry plot is related to the \eqn{K} function (see
\code{\link{Kest}}) and the reduced second moment measure
(see \code{\link{Kmeasure}}). For example, the number
of points in the Fry plot lying within a circle of given radius
is an unnormalised and uncorrected version of the \eqn{K} function.
The Fry plot has a similar appearance to the plot of the
reduced second moment measure \code{\link{Kmeasure}} when the
smoothing parameter \code{sigma} is very small.
The Fry plot does not adjust for the effect
of the size and shape of the sampling window.
The density of points in the Fry plot tapers off near the edges of the
plot. This is an edge effect, a consequence of the bounded sampling
window. In geological applications this is usually not
important, because interest is focused on the behaviour near the
origin where edge effects can be ignored.
To correct for the edge effect, use \code{\link{Kmeasure}} or
\code{\link{Kest}} or its relatives.
}
\value{
\code{fryplot} returns \code{NULL}.
\code{frypoints} returns a point pattern (object of class \code{"ppp"}).
}
\references{
Fry, N. (1979)
Random point distributions and strain measurement in rocks.
\emph{Tectonophysics} \bold{60}, 89--105.
Hanna, S.S. and Fry, N. (1979)
A comparison of methods of strain determination in rocks from
southwest Dyfed (Pembrokeshire) and adjacent areas.
\emph{Journal of Structural Geology} \bold{1}, 155--162.
Patterson, A.L. (1934) A Fourier series method for the determination
of the component of inter-atomic distances in crystals.
\emph{Physics Reviews} \bold{46}, 372--376.
Patterson, A.L. (1935) A direct method for the determination of the
components of inter-atomic distances in crystals.
\emph{Zeitschrift fuer Krystallographie} \bold{90}, 517--554.
}
\seealso{
\code{\link{Kmeasure}},
\code{\link{Kest}}
}
\examples{
## unmarked data
fryplot(cells)
Y <- frypoints(cells)
## numerical marks
fryplot(longleaf, width=4, axes=TRUE)
## multitype points
fryplot(amacrine, width=0.2,
from=(marks(amacrine) == "on"),
chars=c(3,16), cols=2:3,
main="Fry plot centred at an On-cell")
points(0,0)
}
\author{\adrian
and \rolf
}
\keyword{spatial}
\keyword{nonparametric}
Computing file changes ...