Multiple block sizes and overlapping blocks for multivariate time series extremes

DOI10.1214/20-AOS1957MaRDI QIDQ2656597zbMATHOpenAlexFDO

Authors Nan Zou, Stanislav Volgushev, Axel Bücher

Publication date 11 March 2021

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1907.09477

zbMATH Keywords

extreme-value copulaempirical processblock maximamixing coefficientsbias correctionsecond-order condition

Mathematics Subject Classification ID

Asymptotic properties of nonparametric inference (62G20) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Large deviations (60F10) Extreme value theory; extremal stochastic processes (60G70) Asymptotic distribution theory in statistics (62E20) Statistics of extreme values; tail inference (62G32) Characterization and structure theory for multivariate probability distributions; copulas (62H05) Estimation in multivariate analysis (62H12)

Abstract: Block maxima methods constitute a fundamental part of the statistical toolbox in extreme value analysis. However, most of the corresponding theory is derived under the simplifying assumption that block maxima are independent observations from a genuine extreme value distribution. In practice however, block sizes are finite and observations from different blocks are dependent. Theory respecting the latter complications is not well developed, and, in the multivariate case, has only recently been established for disjoint blocks of a single block size. We show that using overlapping blocks instead of disjoint blocks leads to a uniform improvement in the asymptotic variance of the multivariate empirical distribution function of rescaled block maxima and any smooth functionals thereof (such as the empirical copula), without any sacrifice in the asymptotic bias. We further derive functional central limit theorems for multivariate empirical distribution functions and empirical copulas that are uniform in the block size parameter, which seems to be the first result of this kind for estimators based on block maxima in general. The theory allows for various aggregation schemes over multiple block sizes, leading to substantial improvements over the single block length case and opens the door to further methodology developments. In particular, we consider bias correction procedures that can improve the convergence rates of extreme-value estimators and shed some new light on estimation of the second-order parameter when the main purpose is bias correction.

Recommendations

• Extreme value copula estimation based on block maxima of a multivariate stationary time series
• Inference for heavy tailed stationary time series based on sliding blocks
• Weak convergence of a pseudo maximum likelihood estimator for the extremal index
• Maximum likelihood estimators based on the block maxima method
• Asymptotics for sliding blocks estimators of rare events

Cites work

• scientific article; zbMATH DE number 3141621 (Why is no real title available?)
• scientific article; zbMATH DE number 3820920 (Why is no real title available?)
• scientific article; zbMATH DE number 4030574 (Why is no real title available?)
• scientific article; zbMATH DE number 3692406 (Why is no real title available?)
• scientific article; zbMATH DE number 597912 (Why is no real title available?)
• A nonparametric estimation procedure for bivariate extreme value copulas
• A sliding blocks estimator for the extremal index
• Adapting extreme value statistics to financial time series: dealing with bias and serial dependence
• An efficient semiparametric maxima estimator of the extremal index
• Asymptotically distribution-free goodness-of-fit testing for tail copulas
• Asymptotics of empirical copula processes under non-restrictive smoothness assumptions
• Bias correction in multivariate extremes
• Bias-corrected estimation of stable tail dependence function
• Existence and consistency of the maximum likelihood estimators for the extreme value index within the block maxima framework
• Extreme value copula estimation based on block maxima of a multivariate stationary time series
• Extreme value theory for multivariate stationary sequences
• Extreme value theory. An introduction.
• How to make a Hill plot.
• Inference for heavy tailed stationary time series based on sliding blocks
• Limit theorems for empirical processes of cluster functionals
• Limit theory for multivariate sample extremes
• Max-infinite divisibility
• Maximum likelihood estimation for the Fréchet distribution based on block maxima extracted from a time series
• Maximum likelihood estimators based on the block maxima method
• Modelling pairwise dependence of maxima in space
• Multivariate extreme values in stationary random sequences
• Non-parametric Estimation of Tail Dependence
• Nonparametric estimation of multivariate extreme-value copulas
• On second order conditions in the multivariate block maxima and peak over threshold method
• On the block maxima method in extreme value theory: PWM estimators
• On the covariance of the asymptotic empirical copula process
• Rank-based inference for bivariate extreme-value copulas
• Statistics of Extremes
• Statistics of Heteroscedastic Extremes
• Tails of multivariate Archimedean copulas
• Weak convergence and empirical processes. With applications to statistics
• Weak convergence of a pseudo maximum likelihood estimator for the extremal index

Cited in

(16)

• Estimation of cluster functionals for regularly varying time series: sliding blocks estimators
• Estimation of cluster functionals for regularly varying time series: runs estimators
• On second order conditions in the multivariate block maxima and peak over threshold method
• Modelling sub-daily precipitation extremes with the blended generalised extreme value distribution
• Asymptotics for sliding blocks estimators of rare events
• Causal discovery in heavy-tailed models
• Permutation bootstrap and the block maxima method
• Statistical analysis for stationary time series at extreme levels: new estimators for the limiting cluster size distribution
• Estimating POT second-order parameter for bias correction
• A horse race between the block maxima method and the peak-over-threshold approach
• Independent block identification in multivariate time series
• Inference for heavy tailed stationary time series based on sliding blocks
• On the disjoint and sliding block maxima method for piecewise stationary time series
• Extreme value copula estimation based on block maxima of a multivariate stationary time series
• Reweighted madogram-type estimator of Pickands dependence function
• Limit theorems for non-degenerate U-statistics of block maxima for time series

This page was built for publication: Multiple block sizes and overlapping blocks for multivariate time series extremes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2656597)