An Extreme-value Test for Structural Breaks in Spatial Trends

Chenyu Han; Ngai Hang Chan; Chun-Yip Yau

doi:10.5705/ss.202022.0029

An Extreme-value Test for Structural Breaks in Spatial Trends

Chenyu Han, Ngai Hang Chan^*, Chun-Yip Yau

^*Corresponding author for this work

Department of Biostatistics

Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review

Abstract

Non-stationary spatial phenomena are common in various fields such as climate and medical image processing. While many methods examine non-stationary spatial covariance structures, more methods are needed for detecting sudden trend breaks in spatial data. Based on the maximal value of the neighboring discrepancy measurement in the sample space, this paper presents an extreme-value test statistic to detect trend breaks. A simulation-based algorithm is developed to detect breaks in spatial trends at various locations, from which the shape of changing boundaries can be revealed. A simulation study reveals that the test is very effective in detecting structural breaks, especially when they appear at the boundary of the sampling region. Analyses of Australian rainfall and lung tumor data demonstrate the accuracy and wide applicability of the proposed method.

Original language	English
Pages (from-to)	1301-1322
Journal	Statistica Sinica
Volume	35
Issue number	3
Online published	2024
DOIs	https://doi.org/10.5705/ss.202022.0029
Publication status	Published - Jul 2025

Bibliographical note

Research Unit(s) information for this publication is provided by the author(s) concerned.

Funding

This research was supported in part by grants from HKSAR-RGC-GRF Numbers 14308218, 14307921 (Chan), and 14302423, 14302719, 14304221 (Yau).

Research Keywords

Change boundary
Extreme value theory
Inference
Long run variance

Publisher's Copyright Statement

COPYRIGHT TERMS OF DEPOSITED FINAL PUBLISHED VERSION FILE: Statistica Sinica © 2025 Institute of Statistical Science, Academia Sinica. Use of this article is permitted solely for educational and research purposes. Han, C., Chan, N. H., & Yau, C.-Y. (2025). An Extreme-value Test for Structural Breaks in Spatial Trends. Statistica Sinica, 35(3), 1301-1322. https://doi.org/10.5705/ss.202022.0029

Access Output

10.5705/ss.202022.0029

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Embargoed Document

225382316.pdf
Final Published version, 2.17 MB
Embargo ends: 1/08/26

GRF: Statistical Modeling of Big Data Networks
CHAN, N. H. (Principal Investigator / Project Coordinator) & CHEUNG, K. C. (Co-Investigator)
1/01/22 → …
Project: Research

Cite this

@article{e59e5a5b37f14a148698fdd13ec6d6b7,

title = "An Extreme-value Test for Structural Breaks in Spatial Trends",

abstract = "Non-stationary spatial phenomena are common in various fields such as climate and medical image processing. While many methods examine non-stationary spatial covariance structures, more methods are needed for detecting sudden trend breaks in spatial data. Based on the maximal value of the neighboring discrepancy measurement in the sample space, this paper presents an extreme-value test statistic to detect trend breaks. A simulation-based algorithm is developed to detect breaks in spatial trends at various locations, from which the shape of changing boundaries can be revealed. A simulation study reveals that the test is very effective in detecting structural breaks, especially when they appear at the boundary of the sampling region. Analyses of Australian rainfall and lung tumor data demonstrate the accuracy and wide applicability of the proposed method.",

keywords = "Change boundary, Extreme value theory, Inference, Long run variance",

author = "Chenyu Han and Chan, {Ngai Hang} and Chun-Yip Yau",

note = "Research Unit(s) information for this publication is provided by the author(s) concerned.",

year = "2025",

month = jul,

doi = "10.5705/ss.202022.0029",

language = "English",

volume = "35",

pages = "1301--1322",

journal = "Statistica Sinica",

issn = "1017-0405",

publisher = "Academia Sinica * Institute of Statistical Science",

number = "3",

}

TY - JOUR

T1 - An Extreme-value Test for Structural Breaks in Spatial Trends

AU - Han, Chenyu

AU - Chan, Ngai Hang

AU - Yau, Chun-Yip

N1 - Research Unit(s) information for this publication is provided by the author(s) concerned.

PY - 2025/7

Y1 - 2025/7

N2 - Non-stationary spatial phenomena are common in various fields such as climate and medical image processing. While many methods examine non-stationary spatial covariance structures, more methods are needed for detecting sudden trend breaks in spatial data. Based on the maximal value of the neighboring discrepancy measurement in the sample space, this paper presents an extreme-value test statistic to detect trend breaks. A simulation-based algorithm is developed to detect breaks in spatial trends at various locations, from which the shape of changing boundaries can be revealed. A simulation study reveals that the test is very effective in detecting structural breaks, especially when they appear at the boundary of the sampling region. Analyses of Australian rainfall and lung tumor data demonstrate the accuracy and wide applicability of the proposed method.

AB - Non-stationary spatial phenomena are common in various fields such as climate and medical image processing. While many methods examine non-stationary spatial covariance structures, more methods are needed for detecting sudden trend breaks in spatial data. Based on the maximal value of the neighboring discrepancy measurement in the sample space, this paper presents an extreme-value test statistic to detect trend breaks. A simulation-based algorithm is developed to detect breaks in spatial trends at various locations, from which the shape of changing boundaries can be revealed. A simulation study reveals that the test is very effective in detecting structural breaks, especially when they appear at the boundary of the sampling region. Analyses of Australian rainfall and lung tumor data demonstrate the accuracy and wide applicability of the proposed method.

KW - Change boundary

KW - Extreme value theory

KW - Inference

KW - Long run variance

U2 - 10.5705/ss.202022.0029

DO - 10.5705/ss.202022.0029

M3 - RGC 21 - Publication in refereed journal

SN - 1017-0405

VL - 35

SP - 1301

EP - 1322

JO - Statistica Sinica

JF - Statistica Sinica

IS - 3

ER -

An Extreme-value Test for Structural Breaks in Spatial Trends

Abstract

Bibliographical note

Funding

Research Keywords

Publisher's Copyright Statement

Access Output

Other Access Options

Permanent Link

Embargoed Document

Fingerprint

Projects

GRF: Statistical Modeling of Big Data Networks

Cite this