High-dimensional peaks-over-threshold inference
From MaRDI portal
Publication:4561013
DOI10.1093/BIOMET/ASY026zbMATH Open1499.62158arXiv1605.08558OpenAlexW2625415604WikidataQ129977329 ScholiaQ129977329MaRDI QIDQ4561013FDOQ4561013
Authors: Raphaël de Fondeville, A. C. Davison 
Publication date: 10 December 2018
Published in: Biometrika (Search for Journal in Brave)
Abstract: Max-stable processes are increasingly widely used for modelling complex extreme events, but existing fitting methods are computationally demanding, limiting applications to a few dozen variables. -Pareto processes are mathematically simpler and have the potential advantage of incorporating all relevant extreme events, by generalizing the notion of a univariate exceedance. In this paper we investigate score matching for performing high-dimensional peaks over threshold inference, focusing on extreme value processes associated to log-Gaussian random functions and discuss the behaviour of the proposed estimators for regularly-varying distributions with normalized marginals. Their performance is assessed on grids with several hundred locations, simulating from both the true model and from its domain of attraction. We illustrate the potential and flexibility of our methods by modelling extreme rainfall on a grid with locations, based on risks for exceedances over local quantiles and for large spatially accumulated rainfall, and briefly discuss diagnostics of model fit. The differences between the two fitted models highlight the importance of the choice of risk and its impact on the dependence structure.
Full work available at URL: https://arxiv.org/abs/1605.08558
Recommendations
- Multivariate peaks over thresholds models
- Efficient inference for spatial extreme value processes associated to log-Gaussian random functions
- Efficient inference and simulation for elliptical Pareto processes
- High-dimensional inference using the extremal skew-\(t\) process
- Statistical inference for max-stable processes by conditioning on extreme events
proper scoring rulePareto processstatistics of extremesfunctional regular variationgradient scorepeaks-over-threshold analysis
Extreme value theory; extremal stochastic processes (60G70) Statistics of extreme values; tail inference (62G32) Applications of statistics to environmental and related topics (62P12) Stable stochastic processes (60G52)
Cited In (31)
- A Vecchia approximation for high-dimensional Gaussian cumulative distribution functions arising from spatial data
- Modeling Spatial Processes with Unknown Extremal Dependence Class
- Modeling spatial tail dependence with Cauchy convolution processes
- Estimating high-resolution red sea surface temperature hotspots, using a low-rank semiparametric spatial model
- Advances in statistical modeling of spatial extremes
- A modeler's guide to extreme value software
- High-dimensional modeling of spatial and spatio-temporal conditional extremes using INLA and Gaussian Markov random fields
- High-dimensional inference using the extremal skew-\(t\) process
- Multivariate peaks over thresholds models
- Spatial dependence of extreme seas in the North East Atlantic from satellite altimeter measurements
- Max-convolution processes with random shape indicator kernels
- Correlation of powers of Hüsler-Reiss vectors and Brown-Resnick fields, and application to insured wind losses
- Spatial extremes and stochastic geometry for Gaussian-based peaks-over-threshold processes
- Basin-wide spatial conditional extremes for severe ocean storms
- Editorial: EVA 2019 data competition on spatio-temporal prediction of Red Sea surface temperature extremes
- Generalised least squares estimation of regularly varying space-time processes based on flexible observation schemes
- A hierarchical Bayesian non-asymptotic extreme value model for spatial data
- Spatial hierarchical modeling of threshold exceedances using rate mixtures
- Climate extreme event attribution using multivariate peaks-over-thresholds modeling and counterfactual theory
- Modelling extremes of spatial aggregates of precipitation using conditional methods
- A theoretical analysis of the peaking phenomenon in classification
- Testing for changes in the tail behavior of Brown-Resnick Pareto processes
- Implicit max-stable extremal integrals
- A horse race between the block maxima method and the peak-over-threshold approach
- A comparative tour through the simulation algorithms for max-stable processes
- Efficient inference and simulation for elliptical Pareto processes
- Extremal dependence measure for functional data
- Extremes and regular variation
- Space-time trend detection and dependence modeling in extreme event approaches by functional peaks-over-thresholds: application to precipitation in Burkina Faso
- Full likelihood inference for max-stable data
- Local Likelihood Estimation of Complex Tail Dependence Structures, Applied to U.S. Precipitation Extremes
This page was built for publication: High-dimensional peaks-over-threshold inference
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4561013)
