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

Tveten, Martin; Eckley, Idris A.; Fearnhead, Paul

doi:10.1214/21-aoas1508

Cited by 10 publications

(3 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CAPA framework can be extended beyond the off-line anomaly detection in univariate series that we have considered. Early attempts at this include [11,43] which extend CAPA to analyzing multivariate data, while [9] extends CAPA to an online setting.…”

Section: Discussionmentioning

confidence: 99%

A linear time method for the detection of collective and point anomalies

Fisch

Eckley

Fearnhead

2022

Statistical Analysis

Self Cite

View full text Add to dashboard Cite

The challenge of efficiently identifying anomalies in data sequences is an important statistical problem that now arises in many applications. Although there has been substantial work aimed at making statistical analyses robust to outliers, or point anomalies, there has been much less work on detecting anomalous segments, or collective anomalies, particularly in those settings where point anomalies might also occur. In this article, we introduce collective and point anomalies (CAPA), a computationally efficient approach that is suitable when collective anomalies are characterized by either a change in mean, variance, or both, and distinguishes them from point anomalies. Empirical results show that CAPA has close to linear computational cost as well as being more accurate at detecting and locating collective anomalies than other approaches. We demonstrate the utility of CAPA through its ability to detect exoplanets from light curve data from the Kepler telescope and its capacity to detect machine faults from temperature data.

show abstract

Section: Discussionmentioning

confidence: 99%

A linear time method for the detection of collective and point anomalies

Fisch

Eckley

Fearnhead

2022

Statistical Analysis

Self Cite

View full text Add to dashboard Cite

show abstract

“…Here, nonparametric approaches include, for example, detecting changepoints in distribution using hierarchical clustering [25] or a kernel‐based approach [4]. Most current multivariate methods, including parametric approaches, ignore dependencies between the data sequences (see Wang and Samworth [41] and Tveten et al [40] for exceptions).…”

Section: Introductionmentioning

confidence: 99%

Semiparametric detection of changepoints in location, scale, and copula

Agarwal

Eckley

Fearnhead

2023

Statistical Analysis

View full text Add to dashboard Cite

This paper proposes a new method to detect changepoints in the location and scale of univariate data sequences. The proposed method assumes that the data belong to the location‐scale family of distributions and estimate the associated densities nonparametrically. Specifically, the approach does not require knowledge of the functional form of the distribution of the data sequence. As such, the approach can detect changepoints in many distributions. We also propose a new method to detect changes in the location of multivariate sequences, using the marginals and a copula to capture the dependence between variables without the influence of marginal distributions. The performance of the proposed semiparametric approach is contrasted against both other competing nonparametric and Gaussian methods, via simulation studies, as well as applications arising from health and finance.

show abstract

“…Estimating the point at which a structural break occurs in time series models has been extensively researched. We refer the reader to Csörgö and Horváth [1], Bai and Perron [2], Perron [3], Gombay [4], Chen et al [5], Chen [6], Zou et al [7,8], Wang et al [9], Tveten et al [10] for comprehensive surveys.…”

Section: Introductionmentioning

confidence: 99%

Estimation of Multiple Breaks in Panel Data Models Based on a Modified Screening and Ranking Algorithm

Li,

Xiao,

Chen

2023

Symmetry

View full text Add to dashboard Cite

Structural breaks are often encountered in empirical studies with large panels. This paper considers the estimation of multiple breaks in the mean of panel data model based on a modified screening and ranking algorithm. This algorithm satisfies symmetry and is suitable for both cases where the jump size of break points is positive and negative. The break points are first initially screened based on the adaptive Fisher’s statistic, followed by further screening of the break points using the threshold criterion, and finally the final break points are screened using the information criterion. Furthermore, the consistency of the break point estimators is proved. The Monte Carlo simulation results show that the proposed method performs well even if the error terms are serially correlated or cross-sectionally correlated. Finally, two empirical examples illustrate the use of this method.

show abstract

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

Cited by 10 publications

References 36 publications

A linear time method for the detection of collective and point anomalies

A linear time method for the detection of collective and point anomalies

Semiparametric detection of changepoints in location, scale, and copula

Estimation of Multiple Breaks in Panel Data Models Based on a Modified Screening and Ranking Algorithm

Contact Info

Product

Resources

About