Threshold selection and trimming in extremes

DOI10.1007/S10687-020-00385-0zbMATH Open1466.62318arXiv1903.07942OpenAlexW2995259415MaRDI QIDQ2027092FDOQ2027092

Authors: Martin Bladt, J. Beirlant, Hansjörg Albrecher 

Publication date: 21 May 2021

Published in: Extremes (Search for Journal in Brave)

Abstract: We consider removing lower order statistics from the classical Hill estimator in extreme value statistics, and compensating for it by rescaling the remaining terms. Trajectories of these trimmed statistics as a function of the extent of trimming turn out to be quite flat near the optimal threshold value. For the regularly varying case, the classical threshold selection problem in tail estimation is then revisited, both visually via trimmed Hill plots and, for the Hall class, also mathematically via minimizing the expected empirical variance. This leads to a simple threshold selection procedure for the classical Hill estimator which circumvents the estimation of some of the tail characteristics, a problem which is usually the bottleneck in threshold selection. As a by-product, we derive an alternative estimator of the tail index, which assigns more weight to large observations, and works particularly well for relatively lighter tails. A simple ratio statistic routine is suggested to evaluate the goodness of the implied selection of the threshold. We illustrate the favourable performance and the potential of the proposed method with simulation studies and real insurance data.

Full work available at URL: https://arxiv.org/abs/1903.07942

Recommendations

• Threshold selection in extreme value analysis
• Threshold selection in univariate extreme value analysis
• Threshold selection for extremes under a semiparametric model
• scientific article; zbMATH DE number 3917511
• A diagnostic for selecting the threshold in extreme value analysis
• Threshold selection for multivariate heavy-tailed data
• Multiple thresholds in extremal parameter estimation
• Adaptive Threshold Estimation via Extreme Value Theory

zbMATH Keywords

trimmingregular variationthreshold selectionHall class

Mathematics Subject Classification ID

Large deviations (60F10) Statistics of extreme values; tail inference (62G32) Applications of statistics to actuarial sciences and financial mathematics (62P05) Order statistics; empirical distribution functions (62G30)

Cites Work

• Automated threshold selection for extreme value analysis via ordered goodness-of-fit tests with adjustment for false discovery rate
• Reinsurance. Actuarial and statistical aspects
• Statistics of Extremes
• Extreme value theory. An introduction.
• Title not available (Why is that?)
• Adaptive estimates of parameters of regular variation
• A simple general approach to inference about the tail of a distribution
• The bootstrap methodology in statistics of extremes -- choice of optimal sample fraction
• On exponential representations of log-spacings of extreme order statistics
• A new class of semi-parametric estimators of the second order parameter.
• Selecting the optimal sample fraction in univariate extreme value estimation
• How to make a Hill plot.
• A diagnostic for selecting the threshold in extreme value analysis
• Tail Index Estimation for Heavy-Tailed Models: Accommodation of Bias in Weighted Log-Excesses
• Title not available (Why is that?)
• A Sturdy Reduced-Bias Extreme Quantile (VaR) Estimator
• Using a bootstrap method to choose the sample fraction in tail index estimation
• Extended generalised Pareto models for tail estimation
• Using the bootstrap to estimate mean squared error and select smoothing parameter in nonparametric problems
• Kernel estimates of the tail index of a distribution
• A bootstrap-based method to achieve optimality in estimating the extreme-value index
• Tail Index Estimation, Pareto Quantile Plots, and Regression Diagnostics
• Extreme Value Theory and Statistics of Univariate Extremes: A Review
• New Reduced-bias Estimators of a Positive Extreme Value Index
• Generalized sum plots
• On a minimum distance procedure for threshold selection in tail analysis
• Data-adaptive trimming of the Hill estimator and detection of outliers in the extremes of heavy-tailed data
• Ridge regression estimators for the extreme value index
• Combined tail estimation using censored data and expert information

Cited In (11)

• Threshold selection in univariate extreme value analysis
• Data-adaptive trimming of the Hill estimator and detection of outliers in the extremes of heavy-tailed data
• A heuristic adaptive choice of the threshold for bias-corrected Hill estimators
• A diagnostic for selecting the threshold in extreme value analysis
• Improving the traditional Hill estimator in threshold selection in extreme value theory
• Estimation of tail parameters with missing largest observations
• Sequential Monte Carlo samplers to fit and compare insurance loss models
• Reiss and Thomas' automatic selection of the number of extremes
• Trimmed extreme value estimators for censored heavy-tailed data
• An entropy-based validation of threshold selection technique for extreme value analysis and risk assessment
• Title not available (Why is that?)

This page was built for publication: Threshold selection and trimming in extremes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2027092)