Home > Research > Publications & Outputs > Testing for a Change in Mean After Changepoint ...

Electronic data

  • TestingForChange_MainPaper

    Accepted author manuscript, 2.1 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

  • Supplementary Material

    Rights statement: 12m

    Other version, 1.21 MB, PDF document

    Embargo ends: 1/01/50

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License


Text available via DOI:


View graph of relations

Testing for a Change in Mean After Changepoint Detection

Research output: Contribution to Journal/MagazineJournal articlepeer-review

E-pub ahead of print
<mark>Journal publication date</mark>12/04/2022
<mark>Journal</mark>Journal of the Royal Statistical Society: Series B (Statistical Methodology)
Publication StatusE-pub ahead of print
Early online date12/04/22
<mark>Original language</mark>English


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, variants of binary segmentation, $\ell_{0}$ segmentation, or the fused lasso. Our setup allows us to condition on much smaller selection events than existing approaches, which yields higher powered tests. Our procedure leads to improved power in simulation and additional discoveries in a dataset of chromosomal guanine-cytosine content. Our new changepoint inference procedures are freely available in the R package ChangepointInference at https://jewellsean.github.io/changepoint-inference.