Changepoint localization via nested exchangeability tests: from point estimates to confidence intervals
stat.ME, stat.ML
Submitted: 2026-11-05
Updated: 2026-11-05
Comments: 40 pages, 7 figures
License: http://creativecommons.org/licenses/by/4.0/
The gist: We study distribution-free uncertainty quantification for a changepoint estimate in an offline setting.
Terminology
Abstract
We study distribution-free uncertainty quantification for a changepoint estimate in an offline setting. We observe a sequence with a single changepoint and are given an estimate of its location. The pre- and post-change segments are independent and each consists of i.i.d. (or, more generally, exchangeable) observations from arbitrary and unknown distributions. We introduce Changepoint LOcalization via Nested Exchangeability testing (CLONE), a novel method for constructing distribution-free confidence intervals around that estimate by testing exchangeability in nested windows centered at the reference estimate. Under these assumptions, CLONE provides finite-sample coverage for any black-box reference estimator. Using a geometrically shrinking radius grid, CLONE requires only O(log n) permutation tests. We characterize optimal scores for the framework and provide practical recommendations that result in sharp confidence intervals. Under mild assumptions, CLONE confidence intervals have asymptotically vanishing normalized length. Simulations and experiments on image and text data support our theoretical findings.
Sources
- Conformal changepoint localization
- DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
Related papers
- Learning Robust Treatment Rules for Censored Data
- Neural Estimation of Summary Statistics for Simulation-Based Inference with Non-Asymptotic Guarantees
- Spectral divide-and-conquer MCMC for long stationary time series
- Improving the Efficiency of Subgroup Analysis in Randomized Controlled Trials with TMLE
- Personalized Federated Vector Autoregression with Personalization Diversity
- Doubly Robust and Efficient Calibration of Prediction Sets for Right-Censored Time-to-Event Outcomes