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
spokes.Rd
\name{spokes}
\alias{spokes}
\title{Spokes pattern of dummy points}
\description{
Generates a pattern of dummy points in a window,
given a data point pattern.
The dummy points lie on the radii of circles emanating from each
data point.
}
\usage{
spokes(x, y, nrad = 3, nper = 3, fctr = 1.5, Mdefault = 1)
}
\arguments{
\item{x}{
Vector of \eqn{x} coordinates of data points, or a list
with components \code{x} and \code{y}, or a point pattern
(an object of class \code{ppp}).
}
\item{y}{
Vector of \eqn{y} coordinates of data points. Ignored
unless \code{x} is a vector.
}
\item{nrad}{
Number of radii emanating from each data point.
}
\item{nper}{
Number of dummy points per radius.
}
\item{fctr}{
Scale factor.
Length of largest spoke radius is \code{fctr * M}
where \code{M} is the mean nearest neighbour distance
for the data points.
}
\item{Mdefault}{
Value of \code{M} to be used if \code{x} has length 1.
}
}
\value{
If argument \code{x} is a point pattern, a point pattern with
window equal to that of \code{x}. Otherwise a list with two
components \code{x} and \code{y}. In either case the components
\code{x} and \code{y} of the value are numeric vectors giving
the coordinates of the dummy points.
}
\details{
This function is useful in creating dummy points for quadrature
schemes (see \code{\link{quadscheme}}).
Given the data points, the function creates a collection of
\code{nrad * nper * length(x)} dummy points.
Around each data point \code{(x[i],y[i])} there are
\code{nrad * nper} dummy points, lying on \code{nrad} radii
emanating from \code{(x[i],y[i])}, with \code{nper} dummy points
equally spaced along each radius.
The (equal) spacing of dummy points along each radius is
controlled by the factor \code{fctr}.
The distance from a data point to the furthest of its associated
dummy points is \code{fctr * M}
where \code{M} is the mean nearest neighbour distance
for the data points.
If there is only one data point the nearest neighbour distance
is infinite, so the value \code{Mdefault} will be used in place
of \code{M}.
If \code{x} is a point pattern, then the value returned is
also a point pattern, which is clipped to the window
of \code{x}. Hence there may be fewer than
\code{nrad * nper * length(x)} dummy points in the pattern
returned.
}
\seealso{
\code{\link{quad.object}},
\code{\link{quadscheme}},
\code{\link{inside.owin}},
\code{\link{gridcentres}},
\code{\link{stratrand}}
}
\examples{
dat <- runifrect(10)
\dontrun{
plot(dat)
}
dum <- spokes(dat$x, dat$y)
\dontrun{
points(dum$x, dum$y, pch=".")
}
Q <- quadscheme(dat, dum)
}
\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{datagen}
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...