loading page

A Comparison of Multivariate Log Gaussian Cox Process and Saturated Pairwise Interaction Gibbs Point Process
  • Chathuri Samarasekara,
  • Yan Wang,
  • Ian Flint
Chathuri Samarasekara
RMIT University

Corresponding Author:[email protected]

Author Profile
Yan Wang
RMIT University
Author Profile
Ian Flint
The University of Melbourne School of Agriculture Food and Ecosystem Sciences
Author Profile

Abstract

The study of the spatial point patterns in ecology, such as the records of the observed locations of trees, shrubs, nests, burrows, or documented animal presence, relies on multivariate point process models. This study aims to compare the efficacy and applicability of two prominent multivariate point process models, the multivariate log Gaussian Cox process (MLGCP) and the Saturated Pairwise Interaction Gibbs Point Process model (SPIGPP) , highlighting their respective strengths and weaknesses in various scenarios. Using synthetic and real datasets, we assessed both models based on their predictive accuracy of the empirical K function (can we say this?). Our analysis revealed that both MLGCP and SPIGPP effectively identify and capture mild to moderate attractions and regulations. MLGCP struggles to capture repulsive associations as they intensify. In contrast, SPIGPP can well estimates both the direction and magnitude of interactions even when the model is miss-specified. Both models present unique advantages: MLGCP is particularly effective when there is a need to account for complex, unobserved heterogeneities that vary across space, while SPIGPP is suitable when interactions between points are the primary focus. The choice between these models should be guided by the specific needs of the research question and data characteristics.
13 Sep 2024Submitted to Ecology and Evolution
13 Sep 2024Submission Checks Completed
13 Sep 2024Assigned to Editor
26 Sep 2024Reviewer(s) Assigned
14 Oct 2024Review(s) Completed, Editorial Evaluation Pending
15 Oct 2024Editorial Decision: Revise Minor