Applications of Extreme Value Theory to Environmental Data Analysis
Applications of Extreme Value Theory to Environmental Data Analysis
复制标题
极值理论在环境数据分析中的应用
DOI：
10.1002/9781119157052.ch2
复制
发表时间：
2015
影响因子：
3.7
通讯作者：
P. Naveau
中科院分区：
文献类型：
--
作者：
Gwladys Toulemonde;P. Ribéreau;P. Naveau
本章介绍有关单变量和多元极值理论（EVT）的一些基本概念。通过一系列极端数据分析，在环境背景下提出单变量和多元基本概念。这些概念可以通过最大稳定场扩展到空间情况。气候科学是EVT的主要应用领域之一，但也可以包括水文学、金融和保险。本章提供了一系列示例，旨在举例说明概率论如何帮助实践者在多元背景下对极端分位数进行推断。它还讨论了块最大值结果的模拟。对此类过程的推论可以使用 Lindsay 和 Varin 中描述的复合似然性来获得，该应用适用于极值环境，并举例说明了美国极端降水量和年度最大雪深。
This chapter presents some basic concepts about univariate and multivariate extreme value theory (EVT). Through a series of extreme data analysis, univariate and multivariate basic concepts are presented in an environmental context. These concepts can be extended to the spatial case through max‐stable fields. Climate sciences is one of the main fields of applications of EVT, but hydrology, finance, and assurance, can also be included. The chapter presents a series of examples that are treated to exemplify how the probability theory can help the practitioner to make inferences about extreme quantiles within a multivariate context. It also discusses analog of block maxima results. Inference on such processes can be obtained using composite likelihood as described in Lindsay and Varin for an application in an extreme value context with an illustration on US precipitation extremes and for an illustration on annual maximum snow depth.