New Exploratory Tools for Extremal Dependence: χ Networks and Annual Extremal Networks

Whitney K. Huang, Daniel S. Cooley, Imme Ebert-Uphoff, Chen Chen, Snigdhansu Chatterjee

Research output: Contribution to journalArticlepeer-review

5 Scopus citations

Abstract

Understanding dependence structure among extreme values plays an important role in risk assessment in environmental studies. In this work, we propose the χ network and the annual extremal network for exploring the extremal dependence structure of environmental processes. A χ network is constructed by connecting pairs whose estimated upper tail dependence coefficient, χ^ , exceeds a prescribed threshold. We develop an initial χ network estimator, and we use a spatial block bootstrap to assess both the bias and variance of our estimator. We then develop a method to correct the bias of the initial estimator by incorporating the spatial structure in χ. In addition to the χ network, which assesses spatial extremal dependence over an extended period of time, we further introduce an annual extremal network to explore the year-to-year temporal variation of extremal connections. We illustrate the χ and the annual extremal networks by analyzing the hurricane season maximum precipitation at the US Gulf Coast and surrounding area. Analysis suggests there exists long distance extremal dependence for precipitation extremes in the study region and the strength of the extremal dependence may depend on some regional scale meteorological conditions, for example, sea surface temperature.

Original languageEnglish (US)
Pages (from-to)484-501
Number of pages18
JournalJournal of Agricultural, Biological, and Environmental Statistics
Volume24
Issue number3
DOIs
StatePublished - Sep 15 2019

Bibliographical note

Publisher Copyright:
© 2019, International Biometric Society.

Keywords

  • External dependence
  • Hurricanes
  • Networks
  • Precipitation
  • Spatial extremes

Fingerprint

Dive into the research topics of 'New Exploratory Tools for Extremal Dependence: χ Networks and Annual Extremal Networks'. Together they form a unique fingerprint.

Cite this