CFE 2019: Start Registration
View Submission - CMStatistics
Title: Spatio-temporal clustering of extremal behaviour for environmental variables Authors:  Christian Rohrbeck - Lancaster University (United Kingdom) [presenting]
Abstract: To address the need for efficient inference for a range of extreme value problems, spatial pooling of information is the standard approach for marginal tail estimation. A spatial-temporal clustering method is introduced which accounts for both the similarity of the marginal tails and the spatial-temporal dependence structure of the data to determine the appropriate level of pooling. Spatio-temporal dependence is incorporated in two ways: to determine the cluster selection and to account for dependence of the data over sites within a cluster when making the marginal inference. We propose a statistical model for the pairwise extremal dependence which accounts for distance across space and time, and accommodates our belief that sites within the same cluster tend to exhibit a higher degree of dependence than sites in different clusters. We use a Bayesian framework which learns about both the number of clusters and their spatial-temporal structure, and that enables the inference of site-specific marginal distributions of extremes to incorporate uncertainty in the clustering allocation. The approach is illustrated based on the analysis of daily river flow levels in the UK.