rmixsurf: Generate a Poisson process surface object

In sppmix: Modeling Spatial Poisson and Related Point Processes

Description

This function creates a Poisson point process intensity surface modeled as a mixture of normal components, on the given 2d window. The means, covariances and component probabilities are chosen randomly based on parameters passed to the function. The number of components can be either fixed or random.

For examples see

http://faculty.missouri.edu/~micheasa/sppmix/sppmix_all_examples.html#rmixsurf

Usage

rmixsurf(m, lambda, sig0, df, rand_m = FALSE, xlim, ylim, dvec, mu0, Sigma0)

Arguments

m	Number of components of the mixture. If omitted, m is uniformly selected from 1 up to 10.
lambda	Average number of points over the window. If omitted lambda is generated from a Gamma with shape~Unif(1,10) and scale~Unif(50,100).
sig0	Tuning parameter for generating a random matrix from an Inverse Wishart distribution.
df	Degrees of freedom for generating a random matrix from an Inverse Wishart distribution.
rand_m	Request a random number of components. When rand_m = TRUE, the function will randomly choose a number of components from 1:m.
xlim, ylim	Vectors defining the observation window. The component means are sampled uniformly over this window.
dvec	A vector of weights used in the Dirichlet distribution used to sample the mixture probabilities. If the dimension of dvec is not the same as the number of components, then dvec is either truncated to the same dimension or repeated to have dimension m. If missing, a vector of ones is used.
mu0, Sigma0	Mean and covariance matrix for a multivariate normal distribution, used to generate all component means. If mu0 is missing the center of the window is used. If Sigma0 is missing it is set to the identity matrix. If both mu0 and Sigma0 are missing, the component means are generated uniformly over the window of observation.

Value

Object of class intensity_surface.

Author(s)

Sakis Micheas

See Also

plotmix_2d, summary.intensity_surface, plot.intensity_surface

Examples

mixsurf1 <- rmixsurf(m = 3, lambda=100)
summary(mixsurf1)
plot(mixsurf1)
plotmix_2d(mixsurf1)
mixsurf2 <- rmixsurf(m = 5, lambda=200, rand_m = TRUE, ylim = c(-3, 3))
summary(mixsurf2)
plot(mixsurf2)
plotmix_2d(mixsurf2)
mixsurf3 <- rmixsurf(m = 5, lambda=200, rand_m = TRUE, Sigma0=.01*diag(2))
summary(mixsurf3)
plot(mixsurf3)
plotmix_2d(mixsurf3)

sppmix documentation built on Jan. 13, 2021, 10:04 p.m.