Home > Research > Publications & Outputs > Exploiting structure of maximum likelihood esti...

Electronic data

  • MLEStructure

    Rights statement: The final, definitive version of this article has been published in the Journal, Technometrics, ?? (?), 2015, © Informa Plc

    Accepted author manuscript, 613 KB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

Exploiting structure of maximum likelihood estimators for extreme value threshold selection

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published
<mark>Journal publication date</mark>03/2016
<mark>Journal</mark>Technometrics
Issue number1
Volume58
Number of pages11
Pages (from-to)116-126
Publication StatusPublished
Early online date22/01/15
<mark>Original language</mark>English

Abstract

In order to model the tail of a distribution, one has to define the threshold above or below which an extreme value model produces a suitable fit. Parameter stability plots, whereby one plots maximum likelihood estimates of supposedly threshold-independent parameters against threshold, form one of the main tools for threshold selection by practitioners, principally due to their simplicity. However, one repeated criticism of these plots is their lack of interpretability, with pointwise confidence intervals being strongly dependent across the range of thresholds. In this article, we exploit the independent-increments structure of maximum likelihood estimators in order to produce complementary plots with greater interpretability, and a suggest simple likelihood-based procedure which allows for automated threshold selection.

Bibliographic note

The final, definitive version of this article has been published in the Journal, Technometrics, 58 (1), 2016, © Informa Plc