rstpoint {spagmix}

Generate random points in space-time

Description

Generates a random spatiotemporal point pattern containing n independent, identically distributed points with a specified distribution.

Usage

rstpoint(n, f, W = NULL, correction = 1.5, maxpass = 50)

Arguments

n

The number of points to be generated.

f

The probability density of the points, an object of class stim or stden.

W

The polygonal owin defining the spatial window on which the density is defined. If NULL, this will be set to the as.polygonal version of the pixel images stored in f. See ‘Details’.

correction

An adjustment to the number of points generated at the initial pass of the internal loop in an effort to minimise the total number of passes required to reach n points.

maxpass

The maximum number of passes allowed before the function exits. If this is reached before n points are found with respect to the spatiotemporal domain of f, a warning is issued.

Details

This function randomly generates a spatiotemporal point pattern of exactly n points based on the density function f. At any given pass, n * correction points are generated and rejection sampling is used to accept some of the points; this is repeated until the required number of points is found.

The argument W is optional, but is useful when the user wants the spatial window of the resulting point pattern to be a corresponding irregular polygon, as opposed to being based on the boundary of a binary image mask (which, when the pixel images in f are converted to a polygon directly, gives jagged edges based on the union of the pixels).

Value

An object of class ppp containing the n generated points. The marks of the object contain the correspondingly generated observation times.

Author(s)

A.K. Redmond and T.M. Davies

Examples

r1a <- sgmix(cbind(c(0.5,0.5)),vcv=0.01,window=toywin,p0=0.5,p=c(0.5),res=128)
r1b <- sgmix(cbind(c(0.5,0.5),c(0.4,0.6)),vcv=c(0.06,0.015),window=toywin,
             p0=0.1,p=c(0.5,0.4),res=128)
r1c <- sgmix(cbind(c(0.4,0.6)),vcv=c(0.1),window=toywin,p0=0.1,p=c(0.9),res=128)
sts1 <- stkey(start=r1a,
              stop=r1c,
              tlim=c(1,10),
              tres=64,
              window=toywin,
              kf=solist(r1a,r1b),
              kftimes=c(2,6),
              fscale=0.1+0.9*dnorm(seq(-3,3,length=64),mean=0,sd=1))
plot(sts1,sleep=0.1)

Y <- rstpoint(500,sts1,W=toywin,correction=10,maxpass=500)
plot(Y)

require("rgl")
plot3d(Y$x,Y$y,marks(Y))