Time Series Segmentation Procedures to Detect, Locate and Estimate Change-Points

Badagián, Ana Laura; Kaiser, Regina; Peña, Daniel

doi:10.1007/978-3-319-03122-4_3

Cited by 12 publications

(6 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are multiple ways to handle regime changes, such as with time-varying AR, MA, CL, and CLMA terms to reflect parameter changes (Bringmann et al, 2016), while keeping in mind that this makes a model sensitive to noise (Boldea & Hall, 2013;Perron, 2006;Stock & Watson, 2009). As with trends, there is no magic bullet for regime changes, and their existence is often uncertain (Badagián, Kaiser, & Peña, 2015). Our model can account for some regime changes with an occasion effect a t , a time-varying unit effect l t Z i (and covariance c ðxyÞ Z ), and impulse terms c ðyÞ u t , c ðxÞ u t , and c ðxyÞ u t that are free to vary.…”

Section: Threats To Causal Inference: Trends and Regime Changesmentioning

confidence: 99%

From Data to Causes I: Building A General Cross-Lagged Panel Model (GCLM)

Zyphur

Allison

Tay³

et al. 2019

Organizational Research Methods

248

346

View full text Add to dashboard Cite

This is the first paper in a series of two that synthesizes, compares, and extends methods for causal inference with longitudinal panel data in a structural equation modeling (SEM) framework. Starting with a cross-lagged approach, this paper builds a general cross-lagged panel model (GCLM) with parameters to account for stable factors while increasing the range of dynamic processes that can be modeled. We illustrate the GCLM by examining the relationship between national income and subjective well-being (SWB), showing how to examine hypotheses about short-run (via Granger-Sims tests) versus long-run effects (via impulse responses). When controlling for stable factors, we find no short-run or long-run effects among these variables, showing national SWB to be relatively stable, whereas income is less so. Our second paper addresses the differences between the GCLM and other methods. Online Supplementary Materials offer an Excel file automating GCLM input for Mplus (with an example also for Lavaan in R) and analyses using additional data sets and all program input/output. We also offer an introductory GCLM presentation at https://youtu.be/tHnnaRNPbXs . We conclude with a discussion of issues surrounding causal inference.

show abstract

Section: Threats To Causal Inference: Trends and Regime Changesmentioning

confidence: 99%

From Data to Causes I: Building A General Cross-Lagged Panel Model (GCLM)

Zyphur

Allison

Tay³

et al. 2019

Organizational Research Methods

248

346

View full text Add to dashboard Cite

show abstract

“…Otherwise, the detected CP divides the dataset into two sub-datasets and the search is iterated, until no more CPs are detected. An issue with the standard BS is that it overestimates the number of CPs [17].…”

Section: A Training (Off-line) Phasementioning

confidence: 99%

Early Video Content Popularity Detection with Change Point Analysis

Skaperas

Mamatas

Chorti

2018

2018 IEEE Global Communications Conference (GLOBECOM)

View full text Add to dashboard Cite

Video content is responsible for more than 70% of the global IP traffic. Consequently, it is important for content delivery infrastructures to rapidly detect and respond to changes in content popularity dynamics. For flexible and highly adaptive solutions, the capability for a quick response should be driven from early (real-time) and low-complexity content popularity detection schemes. In this paper, we focus on the early and low-complexity detection of video content popularity, which we address as a statistical change point (CP) detection problem. Our proposed methodology estimates in real-time the existence, the number, the magnitude and the direction of changes in the average number of video visits by combining: (i) offline and on-line CP schemes; (ii) an improved measurements window segmentation heuristic for the detection of multiple CPs; and (iii) a variation of the moving average convergence divergence (MACD) indicator to detect the direction of changes. We evaluated the proposed framework using a large database of real youtube video visits. The proposed algorithm is shown to accurately identify CPs and the direction of change in the offline phase. Finally, a few illustrative examples of two variations of the on-line algorithm are also included.

show abstract

“…These proposals can be roughly divided into two classes. One class iteratively searches for one changepoint at a time (Anastasiou & Fryzlewicz, 2019; Badagián et al, 2015; Fryzlewicz, 2014; Olshen et al, 2004; Vostrikova, 1981); the canonical example of this approach is binary segmentation. Another class of proposals simultaneously estimates all changepoints by solving a single optimization problem (Auger & Lawrence, 1989; Fearnhead et al, 2019; Haynes et al, 2017; Hocking et al, 2020; Jackson et al, 2005; Jewell & Witten, 2018; Jewell et al, 2020; Killick et al, 2012; Maidstone et al, 2017; Niu & Zhang, 2012; Tibshirani et al, 2005); examples include

ℓ_{0}

segmentation and the fused lasso.…”

Section: Introductionmentioning

confidence: 99%

Testing for a Change in Mean after Changepoint Detection

Jewell

Fearnhead

Witten

2022

Journal of the Royal Statistical Society Series B: Statistical Methodology

View full text Add to dashboard Cite

While many methods are available to detect structural changes in a time series, few procedures are available to quantify the uncertainty of these estimates post‐detection. In this work, we fill this gap by proposing a new framework to test the null hypothesis that there is no change in mean around an estimated changepoint. We further show that it is possible to efficiently carry out this framework in the case of changepoints estimated by binary segmentation and its variants, ℓ0 segmentation, or the fused lasso. Our setup allows us to condition on much less information than existing approaches, which yields higher powered tests. We apply our proposals in a simulation study and on a dataset of chromosomal guanine‐cytosine content. These approaches are freely available in the R package ChangepointInference at https://jewellsean.github.io/changepoint‐inference/.

show abstract

Time Series Segmentation Procedures to Detect, Locate and Estimate Change-Points

Cited by 12 publications

References 64 publications

From Data to Causes I: Building A General Cross-Lagged Panel Model (GCLM)

From Data to Causes I: Building A General Cross-Lagged Panel Model (GCLM)

Early Video Content Popularity Detection with Change Point Analysis

Testing for a Change in Mean after Changepoint Detection

Contact Info

Product

Resources

About