sim.gap: Simulating point patterns with gaps

In cmjt/gapski: Estimation of parameters for void spaces

Description

Simulates point locations from a point process with deletion areas.

Usage

sim.gap(pars = NULL, d = 2, lims = rbind(c(0, 1), c(0, 1)),
  plot.points = FALSE, plot.empirical = FALSE, return.aunts = FALSE)

Arguments

pars	A named vector of parameter values. Required parameters are D, the density of gaps in the pattern; lambda the density of all points in the domain, before deletion, the parameter R the radius of the spherical gaps in the pattern.
d	a numeric value specifying the number of dimensions the point process is to be simulated in, by default this is 2.
lims	An argument specifying the limits of the domain, for 1D this is a vector, for 2D or greater. By deafault this specifies the unit square.
plot.points	Logical, if TRUE, simulated aunts and niece point locations will be plotted (only for 2D).
plot.empirical	Logical, if TRUE, the empirical Palm intensity is plotted, taken from the nspp package
return.parents	Logical, if TRUE, a named list of matricies is returned of both the observed points and aunt locations

Details

The rchild function may only take a single distributional parameter. If the distribution for the number of children generated by each parent is Poisson, then the native rpois is appropriate, as this distribution has a single parameter. For distributions with two or more parameters, those other than child.par must be hard-coded into rchild. For example, if a Binomial(n, 2, p) is required, then function(n, p) rbinom(n = n, size = 2, prob = p) would be an appropriate function for rchild.

Value

A matrix containing simulated point locations.

cmjt/gapski documentation built on May 13, 2019, 8:44 p.m.