pcfdot.inhom
Inhomogeneous Multitype Pair Correlation Function (Type-i-To-Any-Type)
Estimates the inhomogeneous multitype pair correlation function (from type $i$ to any type) for a multitype point pattern.
- Keywords
- spatial, nonparametric
Usage
pcfdot.inhom(X, i, lambdaI = NULL, lambdadot = NULL, ...,
r = NULL, breaks = NULL,
kernel="epanechnikov", bw=NULL, stoyan=0.15,
correction = c("isotropic", "Ripley", "translate"),
sigma = NULL, varcov = NULL)
Arguments
- X
- The observed point pattern, from which an estimate of the inhomogeneous multitype pair correlation function $g_{i\bullet}(r)$ will be computed. It must be a multitype point pattern (a marked point pattern whose marks are a factor)
- i
- Number or character string identifying the type (mark value)
of the points in
X
from which distances are measured. - lambdaI
- Optional.
Values of the estimated intensity function of the points of type
i
. Either a vector giving the intensity values at the points of typei
, a pixel image (object of class"im"
) giving the i - lambdadot
- Optional.
Values of the estimated intensity function of the point pattern
X
. A numeric vector, pixel image orfunction(x,y)
. - r
- Vector of values for the argument $r$ at which $g_{i\bullet}(r)$ should be evaluated. There is a sensible default.
- breaks
- Optional. An alternative to the argument
r
. Not normally invoked by the user. - kernel
- Choice of smoothing kernel, passed to
density.default
. - bw
- Bandwidth for smoothing kernel, passed to
density.default
. - ...
- Other arguments passed to the kernel density estimation
function
density.default
. - stoyan
- Bandwidth coefficient; see Details.
- correction
- Choice of edge correction.
- sigma,varcov
- Optional arguments passed to
density.ppp
to control the smoothing bandwidth, whenlambdaI
orlambdadot
is estimated by kernel smoothing.
Details
The inhomogeneous multitype (type $i$ to any type) pair correlation function $g_{i\bullet}(r)$ is a summary of the dependence between different types of points in a multitype spatial point process that does not have a uniform density of points.
The best intuitive interpretation is the following: the probability $p(r)$ of finding a point of type $i$ at location $x$ and another point of any type at location $y$, where $x$ and $y$ are separated by a distance $r$, is equal to $$p(r) = \lambda_i(x) lambda(y) g(r) \,{\rm d}x \, {\rm d}y$$ where $\lambda_i$ is the intensity function of the process of points of type $i$, and where $\lambda$ is the intensity function of the points of all types. For a multitype Poisson point process, this probability is $p(r) = \lambda_i(x) \lambda(y)$ so $g_{i\bullet}(r) = 1$.
The command pcfdot.inhom
estimates the inhomogeneous
multitype pair correlation using a modified version of
the algorithm in pcf.ppp
.
If the arguments lambdaI
and lambdadot
are missing or
null, they are estimated from X
by kernel smoothing using a
leave-one-out estimator.
Value
- A function value table (object of class
"fv"
). Essentially a data frame containing the variables r the vector of values of the argument $r$ at which the inhomogeneous multitype pair correlation function $g_{i\bullet}(r)$ has been estimated theo vector of values equal to 1, the theoretical value of $g_{i\bullet}(r)$ for the Poisson process trans vector of values of $g_{i\bullet}(r)$ estimated by translation correction iso vector of values of $g_{i\bullet}(r)$ estimated by Ripley isotropic correction - as required.
See Also
Examples
data(amacrine)
plot(pcfdot.inhom(amacrine, "on", stoyan=0.1), legendpos="bottom")