(5) Robust change detection in the dependence structure of multivariate time series

Daniel Vogel, Roland Fried

Research output: Chapter in Book/Report/Conference proceedingChapter

2 Citations (Scopus)

Abstract

A robust change-point test based on the spatial sign covariance matrix is proposed. A major advantage of the test is its computational simplicity, making it particularly appealing for robust, high-dimensional data analysis. We derive the asymptotic distribution of the test statistic for stationary sequences, which we allow to be near-epoch dependent in probability (P NED) with respect to an α-mixing process. Contrary to the usual L2 near-epoch dependence, this short-range dependence condition requires no moment assumptions, and includes arbitrarily heavy-tailed processes. Further, we give a short review of the spatial sign covariance matrix and compare our test to a similar one based on the sample covariance matrix in a simulation study
Original languageEnglish
Title of host publicationModern Nonparametric, Robust and Multivariate Methods
Subtitle of host publicationFestschrift in Honour of Hannu Oja
EditorsKlaus Nordhausen, Sara Taskinen
PublisherSpringer International Publishing
Pages265-288
Number of pages24
ISBN (Electronic)978-3-319-22404-6
ISBN (Print)978-3-319-22403-9
DOIs
Publication statusPublished - 2015

Fingerprint

Multivariate Time Series
Change Detection
Dependence Structure
Covariance matrix
Sample Covariance Matrix
Stationary Sequences
Mixing Processes
Change Point
High-dimensional Data
Asymptotic distribution
Test Statistic
Simplicity
Data analysis
Simulation Study
Moment
Dependent
Range of data

Keywords

  • GARCH
  • near epoch dependence
  • Oja sign covariance matrix
  • orthogonal invariance
  • spatial sign covariance matrix
  • Tyler matrix

Cite this

Vogel, D., & Fried, R. (2015). (5) Robust change detection in the dependence structure of multivariate time series. In K. Nordhausen, & S. Taskinen (Eds.), Modern Nonparametric, Robust and Multivariate Methods: Festschrift in Honour of Hannu Oja (pp. 265-288). Springer International Publishing. https://doi.org/10.1007/978-3-319-22404-6_16

(5) Robust change detection in the dependence structure of multivariate time series. / Vogel, Daniel; Fried, Roland .

Modern Nonparametric, Robust and Multivariate Methods: Festschrift in Honour of Hannu Oja. ed. / Klaus Nordhausen; Sara Taskinen. Springer International Publishing, 2015. p. 265-288.

Research output: Chapter in Book/Report/Conference proceedingChapter

Vogel, D & Fried, R 2015, (5) Robust change detection in the dependence structure of multivariate time series. in K Nordhausen & S Taskinen (eds), Modern Nonparametric, Robust and Multivariate Methods: Festschrift in Honour of Hannu Oja. Springer International Publishing, pp. 265-288. https://doi.org/10.1007/978-3-319-22404-6_16
Vogel D, Fried R. (5) Robust change detection in the dependence structure of multivariate time series. In Nordhausen K, Taskinen S, editors, Modern Nonparametric, Robust and Multivariate Methods: Festschrift in Honour of Hannu Oja. Springer International Publishing. 2015. p. 265-288 https://doi.org/10.1007/978-3-319-22404-6_16
Vogel, Daniel ; Fried, Roland . / (5) Robust change detection in the dependence structure of multivariate time series. Modern Nonparametric, Robust and Multivariate Methods: Festschrift in Honour of Hannu Oja. editor / Klaus Nordhausen ; Sara Taskinen. Springer International Publishing, 2015. pp. 265-288
@inbook{809fbb8a3058414f92c4222c83caa232,
title = "(5) Robust change detection in the dependence structure of multivariate time series",
abstract = "A robust change-point test based on the spatial sign covariance matrix is proposed. A major advantage of the test is its computational simplicity, making it particularly appealing for robust, high-dimensional data analysis. We derive the asymptotic distribution of the test statistic for stationary sequences, which we allow to be near-epoch dependent in probability (P NED) with respect to an α-mixing process. Contrary to the usual L2 near-epoch dependence, this short-range dependence condition requires no moment assumptions, and includes arbitrarily heavy-tailed processes. Further, we give a short review of the spatial sign covariance matrix and compare our test to a similar one based on the sample covariance matrix in a simulation study",
keywords = "GARCH, near epoch dependence, Oja sign covariance matrix, orthogonal invariance, spatial sign covariance matrix, Tyler matrix",
author = "Daniel Vogel and Roland Fried",
year = "2015",
doi = "10.1007/978-3-319-22404-6_16",
language = "English",
isbn = "978-3-319-22403-9",
pages = "265--288",
editor = "Klaus Nordhausen and Sara Taskinen",
booktitle = "Modern Nonparametric, Robust and Multivariate Methods",
publisher = "Springer International Publishing",

}

TY - CHAP

T1 - (5) Robust change detection in the dependence structure of multivariate time series

AU - Vogel, Daniel

AU - Fried, Roland

PY - 2015

Y1 - 2015

N2 - A robust change-point test based on the spatial sign covariance matrix is proposed. A major advantage of the test is its computational simplicity, making it particularly appealing for robust, high-dimensional data analysis. We derive the asymptotic distribution of the test statistic for stationary sequences, which we allow to be near-epoch dependent in probability (P NED) with respect to an α-mixing process. Contrary to the usual L2 near-epoch dependence, this short-range dependence condition requires no moment assumptions, and includes arbitrarily heavy-tailed processes. Further, we give a short review of the spatial sign covariance matrix and compare our test to a similar one based on the sample covariance matrix in a simulation study

AB - A robust change-point test based on the spatial sign covariance matrix is proposed. A major advantage of the test is its computational simplicity, making it particularly appealing for robust, high-dimensional data analysis. We derive the asymptotic distribution of the test statistic for stationary sequences, which we allow to be near-epoch dependent in probability (P NED) with respect to an α-mixing process. Contrary to the usual L2 near-epoch dependence, this short-range dependence condition requires no moment assumptions, and includes arbitrarily heavy-tailed processes. Further, we give a short review of the spatial sign covariance matrix and compare our test to a similar one based on the sample covariance matrix in a simulation study

KW - GARCH

KW - near epoch dependence

KW - Oja sign covariance matrix

KW - orthogonal invariance

KW - spatial sign covariance matrix

KW - Tyler matrix

U2 - 10.1007/978-3-319-22404-6_16

DO - 10.1007/978-3-319-22404-6_16

M3 - Chapter

SN - 978-3-319-22403-9

SP - 265

EP - 288

BT - Modern Nonparametric, Robust and Multivariate Methods

A2 - Nordhausen, Klaus

A2 - Taskinen, Sara

PB - Springer International Publishing

ER -