Joint second-order parameter estimation for spatio-temporal log-Gaussian Cox processes

Abstract

We propose a new fitting method to estimate the set of second-order parameters for the class of homogeneous spatiotemporal log-Gaussian Cox point processes. With simulations, we show that the proposed minimum contrast procedure, based on the spatio-temporal pair correlation function, provides reliable estimates and we compare the results with the current available methods. Moreover, the proposed method can be used in the case of both separable and non-separable parametric specifications of the correlation function of the underlying Gaussian Random Field. We describe earthquake sequences comparing several Cox model specifications.