Detecting Systematic Anomalies Affecting Systems When Inputs Are Stationary Time Series

Applied Stochastic Models in Business and Industry

54 Pages Posted: 9 Jan 2021 Last revised: 11 Sep 2022

See all articles by Ning Sun

Ning Sun

Agri-Food Analytics Lab; Dalhousie University

Chen Yang

Wuhan University

Ricardas Zitikis

Western University

Date Written: November 20, 2020

Abstract

We develop an anomaly-detection method when systematic anomalies, possibly statistically very similar to genuine inputs, are affecting control systems at the input and/or output stages. The method allows anomaly-free inputs (i.e., those before contamination) to originate from a wide class of random sequences, thus opening up possibilities for diverse applications. To illustrate how the method works on data, and how to interpret its results and make decisions, we analyze several actual time series, which are originally non-stationary but in the process of analysis are converted into stationary. As a further illustration, we provide a controlled experiment with anomaly-free inputs following an ARMA time series model under various contamination scenarios.

Keywords: Control systems, anomaly detection, systematic errors, time series.

Suggested Citation

Sun, Ning and Yang, Chen and Zitikis, Ricardas, Detecting Systematic Anomalies Affecting Systems When Inputs Are Stationary Time Series (November 20, 2020). Applied Stochastic Models in Business and Industry, Available at SSRN: https://ssrn.com/abstract=3734027 or http://dx.doi.org/10.2139/ssrn.3734027

Ning Sun

Agri-Food Analytics Lab ( email )

6100 University Avenue
Halifax, Nova Scotial
Halifax
Canada

Dalhousie University ( email )

6100 University Avenue
Halifax, Nova Scotia
Canada

Chen Yang (Contact Author)

Wuhan University ( email )

Wuhan
China

Ricardas Zitikis

Western University ( email )

1151 Richmond Street
Suite 2
London, Ontario N6A 5B8
Canada

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
84
Abstract Views
674
Rank
523,489
PlumX Metrics