Convergence of U-processes in Hölder spaces with application to robust detection of a changed segment

To detect a changed segment (so called epidemic changes) in a time series, variants of the CUSUM statistic are frequently used. However, they are sensitive to outliers in the data and do not perform well for heavy tailed data, especially when short segments get a high weight in the test statistic. We will present a robust test statistic for epidemic changes based on the Wilcoxon statistic. To study their asymptotic behavior, we prove functional limit theorems for U-processes in Hölder spaces. We also study the finite sample behavior via simulations and apply the statistic to a real data example.

• Publication year

  2019

• Venue

  Statistical Papers

• Publication date

  2019-08-27

• Fields of study

  Mathematics

• Identifiers
  DOI 10.1007/s00362-020-01161-9arXiv 1908.10401
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Studentized U-quantile processes under dependence with applications to change-point analysis.

  2015cited by this paper
• M-procedures for detection of a change under weak dependence

  2014cited by this paper
• Two-sample U-statistic processes for long-range dependent data

  2014cited by this paper
• Testing for Change‐Points in Long‐Range Dependent Time Series by Means of a Self‐Normalized Wilcoxon Test

  2014cited by this paper
• Non‐Parametric Change‐Point Tests for Long‐Range Dependent Data

  2013cited by this paper
• CHANGE-POINT DETECTION UNDER DEPENDENCE BASED ON TWO-SAMPLE U-STATISTICS

  2013cited by this paper
• Limiting distribution for the maximal standardized increment of a random walk

  2012cited by this paper
• The limit distribution of the maximum increment of a random walk with regularly varying jump size distribution

  2010cited by this paper
• Estimators for epidemic alternatives

  2010cited by this paper
• Testing for Change Points in Time Series

  2010cited by this paper
• Multiple change-point estimation with U-statistics

  2010cited by this paper
• Estimating a Changed Segment in a Sample

  2007cited by this paper
• Hölder norm test statistics for epidemic change

  2004cited by this paper
• U-Statistics for Change under Alternatives

  2001cited by this paper
• Sources of nonmonotonic power when testing for a shift in mean of a dynamic time series

  1999cited by this paper
• Using the Generalized Likelihood Ratio Statistic for Sequential Detection of a Change-Point

  1995cited by this paper
• On the power of nonparametric changepoint-tests

  1994cited by this paper
• Testing for change-points with rank and sign statistics

  1994cited by this paper
• Tests for change-points with epidemic alternatives

  1993cited by this paper
• Invariance principles for changepoint problems

  1988cited by this paper
• Approximate Tail Probabilities for the Maxima of Some Random Fields

  1988cited by this paper
• Inference about a change point in experimental neurophysiology

  1986cited by this paper
• A functional central limit theorem for strongly mixing sequences of random variables

  1985cited by this paper
• The cusum test of homogeneity with an application in spontaneous abortion epidemiology.

  1985cited by this paper
• Nonparametric statistical procedures for the changepoint problem

  1984cited by this paper
• Moment bounds for stationary mixing sequences

  1980cited by this paper
• A Non-parametric Approach to the Change-point Problem

  1979cited by this paper
• A Non‐Parametric Approach to the Change‐Point Problem

  1979cited by this paper
• Limiting behavior of U-statistics for stationary, absolutely regular processes

  1976cited by this paper
• A Nonparametric Method for the a Posteriori Detection of the “Disorder” Time of a Sequence of Independent Random Variables

  1976cited by this paper

• Nonparametric changed segment detection in functional data

  2026cites this paper
• Multiple change-points detection based on U-Statistics under weak dependence

  2025cites this paper
• Changed segment tests via norm and Brownian projection

  2025cites this paper
• At the edge of Donsker's Theorem: Asymptotics of multiscale scan statistics

  2025cites this paper
• Power of weighted test statistics for structural change in time series

  2024cites this paper
• Power of Weighted Test Statistics for Structural Change in Time Series

  2023cites this paper
• Robust change-point detection for functional time series based on U-statistics and dependent wild bootstrap

  2022cites this paper
• On change-points tests based on two-samples U-Statistics for weakly dependent observations

  2021cites this paper
• Testing mean changes by maximal ratio statistics

  2021cites this paper
• On change-points tests based on two-samples U-Statistics for weakly dependent observations

  2021cites this paper
• Testing mean changes by maximal ratio statistics

  2021cites this paper
• Editorial for the special issue: Change point detection

  2020cites this paper
• Stein’s method of exchangeable pairs in multivariate functional approximations

  2020cites this paper
• Change-point detection based on weighted two-sample U-statistics

  2020cites this paper
• Functional Convergence of U-processes with Size-Dependent Kernels

  2019cites this paper
• Gibbs posterior convergence and the thermodynamic formalism

  2019cites this paper
• Functional convergence of sequential U-processes with size-dependent kernels

  2019cites this paper