fstrauss.bayes: Bayesian inference for stationary Strauss process

In antiphon/rstrauss: Simulates Strauss point process, in 2D and 3D

Bayesian inference for stationary Strauss process

Description

Using direct approximation of the normalizing constant.

Usage

fstrauss.bayes(
  x,
  R,
  priors = list(beta = c(1, 1e-05), gamma = c(1, 1)),
  lower = c(1e-09, 1e-09),
  upper = c(Inf, 1),
  init,
  ...
)

Arguments

x	a list with elements $x:data.frame of locations (ncol 2 or 3), $bbox: matrix with columns giving window bounds
R	the fixed known interaction range. Use profile pl to infer this.
priors	A list of priors. See details.
lower	Lower limits for searching (beta,gamma).
upper	Upper limits for searching (beta,gamma).
init	Initial values for optim.
...	passed on to approximate_strauss_constant Reduced sample border correction with radius R. We use Gamma(a,b) prior for beta and Beta(c,d) prior for Gamma (funny...) so priors-parameter should be a list with elements $gamma=c(a,b) and $beta=c(c,d). At the moment only the MAP is computed and returned in the $par element. See approximate_strauss_constant for approximation options.

antiphon/rstrauss documentation built on June 2, 2022, 7:19 a.m.