fitStrauss3: Fit Strauss model

In adchSTATS/spatstat3dadd: Additions to spatstat

Description

Function for fitting the Strauss model using logistic regression to 3D data.

Usage

fitStrauss3(X, R, nd, erosion.factor = max(R),
  dum.sim.method = "stratified")

Arguments

X	Three dimensional point pattern of class pp3 contained in a window of class box3.
R	A vector of interaction radii
nd	For dum.sim.method = "Poisson" nd^3 is the expected number of dummy points in window eroded by. For dum.sim.method = "stratified" nd specifies the grid dimension (grid is on the eroded window).
erosion.factor	A number defining how much the window shall be eroded in the border correction. If 0 no correction will be done. The default is max(R).
dum.sim.method	A string indicating which model the dummy point should be simulated from. Options "Poisson" and "stratified"

Details

The method used is only for Gibbs point process models on the exponential family form. For the Strauss process this means that the interaction radius should be fixed. When these requirements are satisfied, a logistic regression is fitted.

Value

A data.frame with length(R) row with variables R, beta, gamma and logLik

Author(s)

Andreas Dyreborg Christoffersen [e-mail]

adchSTATS/spatstat3dadd documentation built on May 20, 2019, 9:26 a.m.