6533b837fe1ef96bd12a32b7
RESEARCH PRODUCT
Locally weighted minimum contrast estimation for spatio-temporal log-Gaussian Cox processes
Nicoletta D’angeloGiada AdelfioJorge Mateusubject
Methodology (stat.ME)FOS: Computer and information sciencesLocal models log-Gaussian Cox processes Minimum contrast Second-order characteristics Spatio-temporal point processesStatistics and ProbabilityComputational MathematicsComputational Theory and MathematicsApplied MathematicsSettore SECS-S/01 - StatisticaStatistics - ComputationStatistics - MethodologyComputation (stat.CO)description
A local version of spatio-temporal log-Gaussian Cox processes is proposed by using Local Indicators of Spatio-Temporal Association (LISTA) functions plugged into the minimum contrast procedure, to obtain space as well as time-varying parameters. The new procedure resorts to the joint minimum contrast fitting method to estimate the set of second-order parameters. This approach has the advantage of being suitable in both separable and non-separable parametric specifications of the correlation function of the underlying Gaussian Random Field. Simulation studies to assess the performance of the proposed fitting procedure are presented, and an application to seismic spatio-temporal point pattern data is shown.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2023-04-01 | Computational Statistics & Data Analysis |