2024/05/30 by J. González, Jose Hermenegildo Ramirez Gonzalez, Antonio Murillo Salas +4
Computer Science · Mathematics · #Gaussian Processes and Bayesian Inference #Target Tracking and Data Fusion in Sensor Networks #Time Series Analysis and Forecasting #math.PR #math.ST #msc:60G15 #msc:62M09 #stat.TH
paper · pdf · doi:10.48550/arxiv.2405.19903
73 pages
openalex publication_date 2024/05/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/30 · arxiv created 2026/07/31 · arxiv updated 2026/08/04
Modeling animal movement is essential for addressing various ecological and biological questions. However, developing an effective predictive model for animal movement is a challenging task. In this paper, we introduce a new family of Gaussian processes, derived from the limiting fluctuations of the rescaled occupation-time process of certain branching particle systems, and study its applicability to real animal movement data. We examine two subfamilies and show that these processes exhibit long-range dependence and covariance functions with logarithmic asymptotic growth. For the exponential subfamily used in the applied analysis, the process is also non-stationary and not intrinsically stationary on compact time intervals. These properties are relevant when dealing with animal trajectories that exhibit strong memory. Finally, we illustrate the practical applicability of the proposed model by analyzing bat movement data.