Scalable change-point and anomaly detection in cross-correlated data with an application to condition monitoring

DOI10.1214/21-AOAS1508zbMATH Open1498.62170arXiv2010.06937OpenAlexW3092730326MaRDI QIDQ2154176FDOQ2154176

Paul Fearnhead, Martin Tveten, Idris A. Eckley

Publication date: 14 July 2022

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

Abstract: Motivated by a condition monitoring application arising from subsea engineering we derive a novel, scalable approach to detecting anomalous mean structure in a subset of correlated multivariate time series. Given the need to analyse such series efficiently we explore a computationally efficient approximation of the maximum likelihood solution to the resulting modelling framework, and develop a new dynamic programming algorithm for solving the resulting Binary Quadratic Programme when the precision matrix of the time series at any given time-point is banded. Through a comprehensive simulation study, we show that the resulting methods perform favourably compared to competing methods both in the anomaly and change detection settings, even when the sparsity structure of the precision matrix estimate is misspecified. We also demonstrate its ability to correctly detect faulty time-periods of a pump within the motivating application.

Full work available at URL: https://arxiv.org/abs/2010.06937

Recommendations

• Novel semi-metrics for multivariate change point analysis and anomaly detection
• scientific article; zbMATH DE number 854964
• Multiscale change point detection for dependent data
• Multiscale change point detection
• Change-point detection in multichannel and distributed systems
• Changepoint Detection in the Presence of Outliers
• High-Dimensional, Multiscale Online Changepoint Detection
• Adaptive Change Point Monitoring for High-Dimensional Data

zbMATH Keywords

outlierscross-correlationchange pointsanomalybinary quadratic programming

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Applications of statistics in engineering and industry; control charts (62P30) Estimation in multivariate analysis (62H12)

Cites Work

• Wild binary segmentation for multiple change-point detection
• Title not available (Why is that?)
• Sparse inverse covariance estimation with the graphical lasso
• Change-point detection in panel data via double CUSUM statistic
• High Dimensional Change Point Estimation via Sparse Projection
• Optimal Detection of Changepoints With a Linear Computational Cost
• Comments on: ``Robust estimation of multivariate location and scatter in the presence of cellwise and casewise contamination
• Changepoint Detection in the Presence of Outliers
• Multiple-Change-Point Detection for High Dimensional Time Series via Sparsified Binary Segmentation
• Change‐point detection in panel data
• Simultaneous discovery of rare and common segment variants
• Uniform change point tests in high dimension
• Subset Multivariate Collective and Point Anomaly Detection
• Sequential multi-sensor change-point detection
• Title not available (Why is that?)
• Common Breaks in Means for Cross‐Correlated Fixed‐T Panel Data
• Detection of Changes in Multivariate Time Series With Application to EEG Data
• Minimax rates in sparse, high-dimensional change point detection
• Joint Structural Break Detection and Parameter Estimation in High-Dimensional Nonstationary VAR Models

Cited In (2)

• A communication-efficient, online changepoint detection method for monitoring distributed sensor networks
• Efficient sparsity adaptive changepoint estimation

This page was built for publication: Scalable change-point and anomaly detection in cross-correlated data with an application to condition monitoring

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2154176)