Anisotropic pair correlation function — pcf

Estimate the anisotropic pair correlation function (2d and 3d), as defined in f 1991, f. 5.2-5.3.

Usage

pcf_anisotropic(
  x,
  r,
  u,
  h,
  stoyan = 0.15,
  lambda,
  lambda_h,
  renormalise = TRUE,
  border = 1,
  divisor = "d",
  ...
)

x: pp, list with $x~coordinates $bbox~bounding box
r: radius vector at which to estimate
u: direction to be converted to angles. Use angle_2_unit to transform from angles.
h: half-widths of epanechnicov kernels, vector of two values, one for ranges and and one for angles.
stoyan: If h not given, use h=c( stoyan/lambda^(1/dim), stoyan*pi) (cf. pcf.ppp-function in spatstat-package).
lambda: optional vector of intensity estimates at points
lambda_h: if lambda missing and lambda_h is given, use this bandwidth in a kernel estimate of lambda(x). Otherwise lambda is set to constant.
renormalise: Scale lambda to align with Campbell formula.
border: Use translation correction? Default=1, yes. Only for cuboidal windows.

The antipode symmetry makes it necessary to compute only on one half of the circle/sphere.