Automated threshold selection for extreme value analysis via ordered goodness-of-fit tests with adjustment for false discovery rate
From MaRDI portal
(Redirected from Publication:133065)
Abstract: Threshold selection is a critical issue for extreme value analysis with threshold-based approaches. Under suitable conditions, exceedances over a high threshold have been shown to follow the generalized Pareto distribution (GPD) asymptotically. In practice, however, the threshold must be chosen. If the chosen threshold is too low, the GPD approximation may not hold and bias can occur. If the threshold is chosen too high, reduced sample size increases the variance of parameter estimates. To process batch analyses, commonly used selection methods such as graphical diagnosis are subjective and cannot be automated, while computational methods may not be feasible. We propose to test a set of thresholds through the goodness-of-fit of the GPD for the exceedances, and select the lowest one, above which the data provides adequate fit to the GPD. Previous attempts in this setting are not valid due to the special feature that the multiple tests are done in an ordered fashion. We apply two recently available stopping rules that control the false discovery rate or familywise error rate to ordered goodness-of-fit tests to automate threshold selection. Various model specification tests such as the Cramer-von Mises, Anderson-Darling, Moran's, and a score test are investigated. The performance of the method is assessed in a large scale simulation study that mimics practical return level estimation. This procedure was repeated at hundreds of sites in the western US to generate return level maps of extreme precipitation.
Recommendations
- Threshold selection in univariate extreme value analysis
- Threshold selection for regional peaks-over-threshold data
- Improved threshold diagnostic plots for extreme value analyses
- Automated selection of r for the r largest order statistics approach with adjustment for sequential testing
- A diagnostic for selecting the threshold in extreme value analysis
Cites work
- scientific article; zbMATH DE number 4159879 (Why is no real title available?)
- scientific article; zbMATH DE number 720689 (Why is no real title available?)
- scientific article; zbMATH DE number 3291701 (Why is no real title available?)
- scientific article; zbMATH DE number 3084686 (Why is no real title available?)
- A flexible extreme value mixture model
- A goodness-of-fit test using Moran's statistic with estimated parameters
- A review of extreme value threshold estimation and uncertainty quantification
- Adaptive false discovery rate control under independence and dependence
- An introduction to statistical modeling of extreme values
- Automated threshold selection for extreme value analysis via ordered goodness-of-fit tests with adjustment for false discovery rate
- Discovering the False Discovery Rate
- Estimating the stable index in order to measure tail thickness: a critique
- Exceedances over high thresholds: a guide to threshold selection
- Extended generalised Pareto models for tail estimation
- How to make a Hill plot.
- Improved threshold diagnostic plots for extreme value analyses
- Inference for Clusters of Extreme Values
- Likelihood-Based Procedures for Threshold Diagnostics and Uncertainty in Extreme Value Modelling
- Linking Pareto-tail kernel goodness-of-fit statistics with tail index at optimal threshold and second order estimation
- On optimising the estimation of high quantiles of a probability distribution
- Rate of convergence for the generalized Pareto approximation of the excesses
- Residual life time at great age
- Sequential selection procedures and false discovery rate control
- Simultaneous and selective inference: Current successes and future challenges
- Some results on tests for Poisson processes
- Statistical inference using extreme order statistics
- The Kumaraswamy GEV distribution
- The control of the false discovery rate in multiple testing under dependency.
- Threshold selection for regional peaks-over-threshold data
- Threshold selection in extreme value analysis
- Using a bootstrap method to choose the sample fraction in tail index estimation
Cited in
(18)- Threshold selection in univariate extreme value analysis
- Invited article by M. Gidea: Extreme events and emergency scales
- Scoring predictions at extreme quantiles
- A modeler's guide to extreme value software
- Using the softplus function to construct alternative link functions in generalized linear models and beyond
- Threshold selection and trimming in extremes
- Discussion of ‘On studying extreme values and systematic risks with nonlinear time series models and tail dependence measures’
- eva
- Automated threshold selection for extreme value analysis via ordered goodness-of-fit tests with adjustment for false discovery rate
- An entropy-based validation of threshold selection technique for extreme value analysis and risk assessment
- Estimation for Extreme Conditional Quantiles of Functional Quantile Regression
- The risk function of the goodness-of-fit tests for tail models
- POT-based estimator of the ruin probability in infinite time for loss models: An application to insurance risk
- Improved threshold diagnostic plots for extreme value analyses
- Threshold selection for regional peaks-over-threshold data
- Automated selection of r for the r largest order statistics approach with adjustment for sequential testing
- Multifractal characteristics on multiple pollution variables in Malaysia
- Extreme tail risk estimation with the generalized Pareto distribution under the peaks-over-threshold framework
This page was built for publication: Automated threshold selection for extreme value analysis via ordered goodness-of-fit tests with adjustment for false discovery rate
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q133065)
