S3T: A score statistic for spatiotemporal change point detection

DOI10.1080/07474946.2020.1826796zbMATH Open1461.62134arXiv1706.05331OpenAlexW3124377368MaRDI QIDQ4965666FDOQ4965666

Authors: Junzhuo Chen, Seong-Hee Kim, Yao Xie 

Publication date: 9 March 2021

Published in: Sequential Analysis (Search for Journal in Brave)

Abstract: We present an efficient score statistic, called the $extsf{S}^{3}extsfT$ statistic, to detect the emergence of a spatially and temporally correlated signal from either fixed-sample or sequential data. The signal may cause a men shift and/or a change in the covariance structure. The score statistic can capture both spatial and temporal structures of the change and hence is particularly powerful in detecting weak signals. The score statistic is computationally efficient and statistically powerful. Our main theoretical contribution are accurate analytical approximations on the false alarm rate of the detection procedures, which can be used to calibrate the threshold analytically. Numerical experiments on simulated and real data demonstrate the good performance of our procedure for solar flame detection and water quality monitoring.

Full work available at URL: https://arxiv.org/abs/1706.05331

Recommendations

• Spatio-temporal change-point modeling
• scientific article; zbMATH DE number 2152212
• Spatial rank-based high-dimensional change point detection via random integration
• Detecting change-points in multidimensional stochastic processes
• Change points detection and parameter estimation for multivariate time series
• Change-detection-assisted multiple testing for spatiotemporal data
• Multiple-Change-Point Detection for High Dimensional Time Series via Sparsified Binary Segmentation
• Change point detection by sparse parameter estimation
• Nonparametric change point detection in multivariate piecewise stationary time series
• A general procedure for change-point detection in multivariate time series

zbMATH Keywords

change point detectionscore statisticsfalse alarm controlspatiotemporal data analysis

Mathematics Subject Classification ID

Nonparametric hypothesis testing (62G10) Inference from spatial processes (62M30) Statistics of extreme values; tail inference (62G32) Sequential statistical analysis (62L10) Applications of statistics to environmental and related topics (62P12)

Cites Work

• Time series: theory and methods.
• The likelihood ratio test for a change-point in simple linear regression
• A moving average approach for spatial statistical models of stream networks
• Sequential analysis. Tests and confidence intervals
• Using the generalized likelihood ratio statistic for sequential detection of a change-point
• Detecting the emergence of a signal in a noisy image
• The statistics of gene mapping
• Spatial statistics and modeling. Translated from the French by Kevin Bleakley.
• Title not available (Why is that?)
• Extremes in random fields. A theory and its applications
• Multivariate Generalizations of Cumulative Sum Quality-Control Schemes
• Approximate tail probabilities for the maxima of some random fields
• Title not available (Why is that?)
• Estimating and modeling spatio-temporal correlation structures for river monitoring networks

Cited In (8)

• Detection of change in the spatiotemporal mean function
• Online score statistics for detecting clustered change in network point processes
• Sequential change-point detection: computation versus statistical performance
• Detection of spatiotemporal changepoints: a generalised additive model approach
• Adaptive resources allocation CUSUM for binomial count data monitoring with application to COVID-19 hotspot detection
• Pesticide concentration monitoring: investigating spatio-temporal patterns in left censored data
• Detecting changes in spatial-temporal image data based on quadratic forms
• Sequential detection of three-dimensional signals under dependent noise

Uses Software

• SWMM5

This page was built for publication: S3T: A score statistic for spatiotemporal change point detection

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4965666)