Detecting relevant changes in the mean of nonstationary processes -- a mass excess approach

DOI10.1214/19-AOS1811zbMATH Open1435.62320arXiv1801.09874MaRDI QIDQ2284384FDOQ2284384

Authors: Weichi Wu, Holger Dette 

Publication date: 15 January 2020

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

Abstract: This paper considers the problem of testing if a sequence of means $(m{u}_t{)}_{t=1,ldots,n}$ of a non-stationary time series $(X_t{)}_{t=1,ldots,n}$ is stable in the sense that the difference of the means $m{u}_1$ and $m{u}_t$ between the initial time $t=1$ and any other time is smaller than a given level, that is $|m{u}_1-m{u}_t|leqc$ for all $t=1,ldots,n$. A test for hypotheses of this type is developed using a biascorrected monotone rearranged local linear estimator and asymptotic normality of the corresponding test statistic is established. As the asymptotic variance depends on the location and order of the critical roots of the equation $|m{u}_1-m{u}_t|=c$ a new bootstrap procedure is proposed to obtain critical values and its consistency is established. As a consequence we are able to quantitatively describe relevant deviations of a non-stationary sequence from its initial value. The results are illustrated by means of a simulation study and by analyzing data examples.

Full work available at URL: https://arxiv.org/abs/1801.09874

Recommendations

• A distribution free test for changes in the trend function of locally stationary processes
• Sequential monitoring for changes from stationarity to mild non-stationarity
• Are deviations in a gradually varying mean relevant? A testing approach based on sup-norm estimators
• Detecting relevant changes in time series models
• Consistent nonparametric change point detection combining CUSUM and marked empirical processes

zbMATH Keywords

Gaussian approximationlocal linear estimationlocally stationary processchange point analysisrearrangement estimatorsrelevant change points

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Nonparametric hypothesis testing (62G10) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Nonparametric statistical resampling methods (62G09) General second-order stochastic processes (60G12)

Cites Work

• A simple nonparametric estimator of a strictly monotone regression function
• Title not available (Why is that?)
• Multiscale local change point detection with applications to value-at-risk
• Statistical inference for time-inhomogeneous volatility models.
• Detecting relevant changes in the mean of nonstationary processes -- a mass excess approach
• Improving point and interval estimators of monotone functions by rearrangement
• Quantile and probability curves without crossing
• Non-Crossing Non-Parametric Estimates of Quantile Curves
• Break detection in the covariance structure of multivariate time series models
• Fitting time series models to nonstationary processes
• On nonparametric estimation of density level sets
• Total error in a plug-in estimator of level sets.
• Asymptotic normality of plug-in level set estimates
• Structural breaks in time series
• Tests for Parameter Instability and Structural Change With Unknown Change Point
• PLUG-IN ESTIMATION OF GENERAL LEVEL SETS
• Estimating and Testing Linear Models with Multiple Structural Changes
• Inference of Trends in Time Series
• SLEX Analysis of Multivariate Nonstationary Time Series
• CONTINUOUS INSPECTION SCHEMES
• Asymptotics and optimal bandwidth selection for highest density region estimation
• Kernel estimation of density level sets
• Tests of Equality Between Sets of Coefficients in Two Linear Regressions
• The dip test of unimodality
• Measuring mass concentrations and estimating density contour clusters -- An excess mass approach
• Inference for single and multiple change-points in time series
• Excess Mass Estimates and Tests for Multimodality
• Testing Statistical Hypotheses of Equivalence and Noninferiority
• Banding sample autocovariance matrices of stationary processes
• Calibrating the Excess Mass and Dip Tests of Modality
• Generalized density clustering
• Nonparametric regression for locally stationary time series
• Local linear quantile estimation for nonstationary time series
• Jump-preserving regression and smoothing using local linear fitting: a compromise
• Discrimination of Locally Stationary Time Series Based on the Excess Mass Functional
• Title not available (Why is that?)
• Testing for Structural Change in Dynamic Models
• A jump-preserving curve fitting procedure based on local piecewise-linear kernel estimation
• Trimmed Comparison of Distributions
• Exact Results for Shewhart Control Charts with Supplementary Runs Rules
• Improvement of Kernel Type Density Estimators
• Similarity of samples and trimming
• Simultaneous Inference of Linear Models with Time Varying Coefficients
• Sojourn times
• Detecting relevant changes in time series models
• Estimation of regression contour clusters -- an application of the excess mass approach to regression

Cited In (25)

• Functional Estimation and Change Detection for Nonstationary Time Series
• Detecting long-range dependence for time-varying linear models
• Box-constrained monotone approximations to Lipschitz regularizations, with applications to robust testing
• Detecting relevant changes in the mean of nonstationary processes -- a mass excess approach
• Inference for high‐dimensional linear models with locally stationary error processes
• Distribution of Distances based Object Matching: Asymptotic Inference
• Identifying shifts between two regression curves
• Weighted Dickey-Fuller processes for detecting stationarity
• A distribution free test for changes in the trend function of locally stationary processes
• Detecting changes in the trend function of heteroscedastic time series
• A Scale‐space Approach for Detecting Non‐stationarities in Time Series
• Title not available (Why is that?)
• Detecting relevant changes in time series models
• Optimal difference-based variance estimators in time series: a general framework
• Sequential monitoring for changes from stationarity to mild non-stationarity
• Empirical optimal transport under estimated costs: distributional limits and statistical applications
• The use of temporally aggregated data on detecting a mean change of a time series process
• Modelling time-varying first and second-order structure of time series via wavelets and differencing
• Rearranged dependence measures
• Are deviations in a gradually varying mean relevant? A testing approach based on sup-norm estimators
• Comparing time varying regression quantiles under shift invariance
• Monitoring procedures for strict stationarity based on the multivariate characteristic function
• Detecting changes in mean in the presence of time-varying autocovariance
• Mean stationarity test in time series: a signal variance-based approach
• Prediction in Locally Stationary Time Series

This page was built for publication: Detecting relevant changes in the mean of nonstationary processes -- a mass excess approach

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2284384)