A logistic regression estimating function for spatial Gibbs point processes



One of the most popular estimation techniques for spatial Gibbs point process models is the method of maximum pseudo likelihood (MPL).
Calculation of the MPL estimate requires the evaluation of a spatial integral, which is usually done numerically. However, the numerical integration introduces a bias in the parameter estimates which can be hard to quantify. In this project we are developing an alternative estimation method based on logistic regression which provides unbiased estimates and is computationally simple to handle.
Effektiv start/slut dato01/09/201231/12/2014