Relevant change points in high dimensional time series

DOI10.1214/18-EJS1464MaRDI QIDQ1786570zbMATH OpenOpenAlexFDO

Authors Josua Gösmann, Holger Dette

Publication date 24 September 2018

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1704.04614, https://projecteuclid.org/euclid.ejs/1535681028

zbMATH Keywords

CUSUMhigh-dimensional time seriesphysical dependence measurechange point analysisprecise hypothesesrelevant changes

Mathematics Subject Classification ID

Nonparametric hypothesis testing (62G10) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Asymptotic properties of parametric tests (62F05) Non-Markovian processes: hypothesis testing (62M07)

Abstract: This paper investigates the problem of detecting relevant change points in the mean vector, say $m{u}_t=(m{u}_{1,t},ldots,m{u}_{d,t}{)}^T$ of a high dimensional time series $(Z_t{)}_{tinmathbbZ}$. While the recent literature on testing for change points in this context considers hypotheses for the equality of the means $m{u}_h^{(1)}$ and $m{u}_h^{(2)}$ before and after the change points in the different components, we are interested in a null hypothesis of the form ${\displaystyle }$ H_0: |mu^{(1)}_{h} - mu^{(2)}_{h} | leq Delta_h ~~~mbox{ for all } ~~h=1,ldots ,d ${\displaystyle }$ where $Delt{a}_1,ldots,Delt{a}_d$ are given thresholds for which a smaller difference of the means in the $h$-th component is considered to be non-relevant. We propose a new test for this problem based on the maximum of squared and integrated CUSUM statistics and investigate its properties as the sample size $n$ and the dimension $d$ both converge to infinity. In particular, using Gaussian approximations for the maximum of a large number of dependent random variables, we show that on certain points of the boundary of the null hypothesis a standardised version of the maximum converges weakly to a Gumbel distribution.

Recommendations

• Detecting relevant changes in time series models
• Sequential change point detection in high dimensional time series
• Uniform change point tests in high dimension
• Testing and estimating change-points in the covariance matrix of a high-dimensional time series
• Robust inference for change points in high dimension

Cites work

• A Simple, Positive Semi-Definite, Heteroskedasticity and Autocorrelation Consistent Covariance Matrix
• Banding sample autocovariance matrices of stationary processes
• CONTINUOUS INSPECTION SCHEMES
• CONTROL CHARTS WITH WARNING LINES
• CUSUM-type testing for changing parameters in a spatial autoregressive model for stock returns
• Change-point detection in panel data
• Common breaks in means and variances for panel data
• Detecting relevant changes in time series models
• Detecting simultaneous change points in multiple sequences
• Detection of changes in multivariate time series with application to EEG data
• Gaussian approximation for high dimensional vector under physical dependence
• Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors
• High dimensional change point estimation via sparse projection
• Limit Theorems for the Maximum Term in Stationary Sequences
• Monitoring Structural Change
• Multiple-Change-Point Detection for High Dimensional Time Series via Sparsified Binary Segmentation
• Nonlinear system theory: Another look at dependence
• Resampling methods for dependent data
• Sequential block bootstrap in a Hilbert space with application to change point analysis
• Some Difficulties of Interpretation Encountered in the Application of the Chi-Square Test
• Strong invariance principles for dependent random variables
• Structural breaks in time series
• Testing Statistical Hypotheses of Equivalence and Noninferiority
• Testing for changes in multivariate dependent observations with an application to temperature changes
• The jackknife and the bootstrap for general stationary observations
• Uniform change point tests in high dimension

Cited in

(12)

• Estimating a Change Point in a Sequence of Very High-Dimensional Covariance Matrices
• Inference for change points in high-dimensional data via selfnormalization
• Detecting relevant changes in time series models
• High dimensional efficiency with applications to change point tests
• Sequential change point detection in high dimensional time series
• Frequency Detection and Change Point Estimation for Time Series of Complex Oscillation
• Sequential Gaussian approximation for nonstationary time series in high dimensions
• Detecting Abrupt Changes in High-Dimensional Self-Exciting Poisson Processes
• Detecting changes in the second moment structure of high-dimensional sensor-type data in a \(K\)-sample setting
• Validating approximate slope homogeneity in large panels
• Uniform change point tests in high dimension
• Optimal multiple change-point detection for high-dimensional data

This page was built for publication: Relevant change points in high dimensional time series

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1786570)