On new families of anisotropic spatial log-Gaussian Cox processes
- PDF / 4,872,014 Bytes
- 31 Pages / 595.276 x 790.866 pts Page_size
- 11 Downloads / 297 Views
ORIGINAL PAPER
On new families of anisotropic spatial log-Gaussian Cox processes Fariba Nasirzadeh1 • Zohreh Shishebor1
•
Jorge Mateu2
Accepted: 12 October 2020 Ó Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Cox processes are natural models for point process phenomena that are environmentally driven, but much less natural for phenomena driven primarily by interactions amongst the points. The class of log-Gaussian Cox processes (LGCPs) has an elegant simplicity, and one of its attractive features is the tractability of the multivariate normal distribution carries over, to some extent, to the associated Cox process. In the statistical analysis of spatial point patterns, it is often assumed isotropy because of a simpler interpretation and ease of analysis. However, there are many cases in which the spatial structure depends on the direction. In this paper, we introduce new families of anisotropic spatial LGCPs that are useful to model spatial anisotropic point patterns that exhibit a degree of clustering. We propose classes of families consisting of elliptical and non-elliptical models. The former can be reduced to isotropic forms after some rotations, while the latter family goes beyond this property. We derive analytical forms for the covariance of the associated random field, and some second-order characteristics. A moment-based estimation procedure is followed to make inference on the parameters that control the degree of anisotropy. The estimation procedure is evaluated through a simulation study under a variety of scenarios and various degrees of anisotropy. Our methodology is illustrated on two real datasets of earthquakes in South America and the Mediterranean Europe. Keywords Anisotropy Intensity function K-function Log-Gaussian Cox processes Minimum contrast estimation Pair correlation function Super-ellipse
1 Introduction An outstanding class of stochastic point processes is the class of spatial point processes, defined as random mechanisms to generate a countable set of points randomly located on, usually, a planar space. These processes are applied in many different fields such as geology, seismology, economics, image processing, ecology, or biology; see, as some Examples, Funwi-Gabga and Mateu (2012), Uria et al. (2013), Serra et al. (2014). The Poisson point process is the most basic and simplest model of point processes. This model can be used to build a more flexible and fundamental class of spatial models named Cox processes. A Cox process is obtained as an extension of a & Zohreh Shishebor [email protected] 1
Department of Statistics, Shiraz University, Shiraz, Iran
2
Department of Mathematics, University Jaume I, Castellon, Spain
Poisson process by considering the intensity function of the Poisson process a realization of a random field. Cox processes are natural models for point process phenomena that are environmentally driven, but much less natural for phenomena driven primarily by interactions amongst the points. The simplicity of using a Cox pro
Data Loading...
