Multivariate max-stable spatial processes

• 成果类型: Article
• 署名作者: Genton, Marc G.; Padoan, Simone A.; Sang, Huiyan
• 署名单位: King Abdullah University of Science & Technology; Bocconi University; Texas A&M University System; Texas A&M University College Station
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/asu066
• 发表日期: 2015
• 页码: 215230
• 关键词: cross-covariance functions; composite likelihood; random-fields; extreme values; random vectors; dependence; inference

摘要

Max-stable processes allow the spatial dependence of extremes to be modelled and quantified, so they are widely adopted in applications. For a better understanding of extremes, it may be useful to study several variables simultaneously. To this end, we study the maxima of independent replicates of multivariate processes, both in the Gaussian and Student-t cases. We define a Poisson process construction and introduce multivariate versions of the Smith Gaussian extreme-value, the Schlather extremal-Gaussian and extremal-t, and the Brown-Resnick models. We develop inference for the models based on composite likelihoods. We present results of Monte Carlo simulations and an application to daily maximum wind speed and wind gust.