Home > Research > Publications & Outputs > Detection of Emergent Anomalous Structure in Fu...

Links

Text available via DOI:

View graph of relations

Detection of Emergent Anomalous Structure in Functional Data

Research output: Contribution to Journal/MagazineJournal articlepeer-review

E-pub ahead of print

Standard

Detection of Emergent Anomalous Structure in Functional Data. / Austin, Edward; Eckley, Idris A.; Bardwell, Lawrence.
In: Technometrics, 14.05.2024, p. 1-11.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Austin E, Eckley IA, Bardwell L. Detection of Emergent Anomalous Structure in Functional Data. Technometrics. 2024 May 14;1-11. Epub 2024 May 14. doi: 10.1080/00401706.2024.2342315

Author

Bibtex

@article{c28a6e822ed043c59d637c18de988e14,
title = "Detection of Emergent Anomalous Structure in Functional Data",
abstract = "Motivated by an example arising from digital networks, we propose a novel approach for detecting the emergence of anomalies in functional data. In contrast to classical functional data approaches, which detect anomalies in completely observed curves, the proposed approach seeks to identify anomalies sequentially as each point on the curve is received. The new method, the Functional Anomaly Sequential Test (FAST), captures the common profile of the curves using Principal Differential Analysis and uses a form of CUSUM test to monitor a new functional observation as it emerges. Various theoretical properties of the procedure are derived. The performance of FAST is then assessed on both simulated and telecommunications data.",
author = "Edward Austin and Eckley, {Idris A.} and Lawrence Bardwell",
year = "2024",
month = may,
day = "14",
doi = "10.1080/00401706.2024.2342315",
language = "English",
pages = "1--11",
journal = "Technometrics",
issn = "0040-1706",
publisher = "American Statistical Association",

}

RIS

TY - JOUR

T1 - Detection of Emergent Anomalous Structure in Functional Data

AU - Austin, Edward

AU - Eckley, Idris A.

AU - Bardwell, Lawrence

PY - 2024/5/14

Y1 - 2024/5/14

N2 - Motivated by an example arising from digital networks, we propose a novel approach for detecting the emergence of anomalies in functional data. In contrast to classical functional data approaches, which detect anomalies in completely observed curves, the proposed approach seeks to identify anomalies sequentially as each point on the curve is received. The new method, the Functional Anomaly Sequential Test (FAST), captures the common profile of the curves using Principal Differential Analysis and uses a form of CUSUM test to monitor a new functional observation as it emerges. Various theoretical properties of the procedure are derived. The performance of FAST is then assessed on both simulated and telecommunications data.

AB - Motivated by an example arising from digital networks, we propose a novel approach for detecting the emergence of anomalies in functional data. In contrast to classical functional data approaches, which detect anomalies in completely observed curves, the proposed approach seeks to identify anomalies sequentially as each point on the curve is received. The new method, the Functional Anomaly Sequential Test (FAST), captures the common profile of the curves using Principal Differential Analysis and uses a form of CUSUM test to monitor a new functional observation as it emerges. Various theoretical properties of the procedure are derived. The performance of FAST is then assessed on both simulated and telecommunications data.

U2 - 10.1080/00401706.2024.2342315

DO - 10.1080/00401706.2024.2342315

M3 - Journal article

SP - 1

EP - 11

JO - Technometrics

JF - Technometrics

SN - 0040-1706

ER -