2018/03/18 by Sabrina Vettori, Raphaël Huser, Vettori, Sabrina +3 · 1 citation
Economics, Econometrics and Finance · Environmental Science · #Air Quality Monitoring and Forecasting #Air Quality and Health Impacts #Applications (stat.AP) #FOS: Computer and information sciences #Spatial and Panel Data Analysis
paper · pdf · doi:10.48550/arxiv.1804.04588
openalex publication_date 2018/03/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Capturing the potentially strong dependence among the peak concentrations of\nmultiple air pollutants across a spatial region is crucial for assessing the\nrelated public health risks. In order to investigate the multivariate spatial\ndependence properties of air pollution extremes, we introduce a new class of\nmultivariate max-stable processes. Our proposed model admits a hierarchical\ntree-based formulation, in which the data are conditionally independent given\nsome latent nested \α-stable random factors. The hierarchical structure\nfacilitates Bayesian inference and offers a convenient and interpretable\ncharacterization. We fit this nested multivariate max-stable model to the\nmaxima of air pollution concentrations and temperatures recorded at a number of\nsites in the Los Angeles area, showing that the proposed model succeeds in\ncapturing their complex tail dependence structure.\n