A comparison of single and multiple changepoint techniques for time series data

Shi, Xueheng; Gallagher, Colin; Lund, Robert; Killick, Rebecca

doi:10.1016/j.csda.2022.107433

Cited by 29 publications

(13 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus the test is invalid and will produce spurious false positives. This is well known in the univariate changepoint literature (Shi et al, 2021) and can be mitigated by scaling the threshold by the autocorrelation observed.…”

Section: Finite Sample Propertiesmentioning

confidence: 99%

Detecting changes in covariance via random matrix theory

Ryan¹,

Killick²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

A novel method is proposed for detecting changes in the covariance structure of moderate dimensional time series. This non-linear test statistic has a number of useful properties. Most importantly, it is independent of the underlying structure of the covariance matrix. We discuss how results from Random Matrix Theory, can be used to study the behaviour of our test statistic in a moderate dimensional setting (i.e. the number of variables is comparable to the length of the data). In particular, we demonstrate that the test statistic converges point wise to a normal distribution under the null hypothesis. We evaluate the performance of the proposed approach on a range of simulated datasets and find that it outperforms a range of alternative recently proposed methods. Finally, we use our approach to study changes in the amount of water on the surface of a plot of soil which feeds into model development for degradation of surface piping.

show abstract

Section: Finite Sample Propertiesmentioning

confidence: 99%

Detecting changes in covariance via random matrix theory

Ryan¹,

Killick²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…discussed in [11]. The term |m − k| accounts for the discrepancy between the changepoint counts in the two configurations and the "minimum component" min A(C 1 , C 2 ) measures how well the two changepoint time sets align with one and other.…”

Section: Distances Between Changepoint Configurationsmentioning

confidence: 99%

“…While WBS2-SDLL works better than WBS, performance issues remain. We refer to [11] for additional comparisons of multiple changepoint techniques. Our overarching point is that a more extensive comparison of these techniques is needed in low-and mid-frequency changepoint settings.…”

Section: Wbs Performance In Infrequent Changepoint Settingsmentioning

confidence: 99%

Short Communication: Detecting Possibly Frequent Change-points: Wild Binary Segmentation 2

Lund¹,

Shi²

2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Killick et al, 2012;Maidstone et al, 2017); and for different types of data, such as high-dimensional data (Wang and Samworth, 2018), network data (Wang et al, 2021), and general non-Euclidean data (Song and Chen, 2022;Dubey and Müller, 2020). See Truong et al (2020), Fearnhead and Rigaill (2020) and Shi et al (2022b) for an overview of this area.…”

Section: Introductionmentioning

confidence: 99%

Improving Power by Conditioning on Less in Post-selection Inference for Changepoints

Carrington¹,

Fearnhead²

2023

Preprint

View full text Add to dashboard Cite

Post-selection inference has recently been proposed as a way of quantifying uncertainty about detected changepoints. The idea is to run a changepoint detection algorithm, and then re-use the same data to perform a test for a change near each of the detected changes. By defining the p-value for the test appropriately, so that it is conditional on the information used to choose the test, this approach will produce valid p-values. We show how to improve the power of these procedures by conditioning on less information. This gives rise to an ideal selective p-value that is intractable but can be approximated by Monte Carlo. We show that for any Monte Carlo sample size, this procedure produces valid p-values, and empirically that noticeable increase in power is possible with only very modest Monte Carlo sample sizes.Our procedure is easy to implement given existing post-selection inference methods, as we just need to generate perturbations of the data set and re-apply the post-selection method to each of these. On genomic data consisting of human GC content, our procedure increases the number of significant changepoints that are detected from e.g. 17 to 27, when compared to the method of Jewell et al. (2021).

show abstract

A comparison of single and multiple changepoint techniques for time series data

Cited by 29 publications

References 41 publications

Detecting changes in covariance via random matrix theory

Detecting changes in covariance via random matrix theory

Short Communication: Detecting Possibly Frequent Change-points: Wild Binary Segmentation 2

Improving Power by Conditioning on Less in Post-selection Inference for Changepoints

Contact Info

Product

Resources

About