Announced Breaks Separate Conformal Reliability from Frequency Calibration
Abstract
Many structural breaks are announced in advance, yet adaptive conformal methods learn of them only from their own errors. A reset at the announced date serves one of two audits, conformal reliability of the next interval or frequency calibration of a closed window, and the two can disagree. For independent scores, restarting the empirical quantile at the boundary attains the minimax reliability rate, of order one over the square root of the post-break count. Restarting an adaptive recursion attains a frequency gap of order one over that count on bounded scores when its floor adds a bounded total. That gap bounds nothing about the next interval. After a large synthetic break, the restarted quantile's coverage error at the 50th forecast is 0.035, against 0.052 for the best of five adaptive baselines. The restarted recursion has the smallest frequency gap over the first 50 forecasts after every synthetic break. Real breaks are mostly smaller. The restarted quantile beats two baselines after an electricity market reform and trails all five after monetary policy decisions. Of nine announced breaks screened, only one is large enough for a detectable gain, and there every method misses nominal coverage by more than 0.1.
Related
- Reliability-Centered Evaluation of Sparse Longitudinal CT Lesion-Size Forecasting with Conformal Interval Calibration and Gompertz-Inspired Regularization
- When Does Trimming Help Conformal Prediction? A Retained-Law Diagnostic under Calibration Contamination
- Conformal Calibration Transfer
- Multi-Variable Conformal Prediction: Optimizing Prediction Sets using Multiple Calibration Variables
- When does conformal calibration need censoring weights? Cause-of-failure prediction sets under competing risks
BibTeX
@misc{li2026announced,
title = {{Announced Breaks Separate Conformal Reliability from Frequency Calibration}},
author = {Karl Li},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.23917},
eprint = {2610.23917},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.23917},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Announced Breaks Separate Conformal Reliability from Frequency Calibration
AU - Karl Li
PY - 2026
DA - 2026/10/26
PB - arXiv
DO - 10.48550/arxiv.2610.23917
UR - https://arxiv.org/abs/2610.23917
AB - Many structural breaks are announced in advance, yet adaptive conformal methods learn of them only from their own errors. A reset at the announced date serves one of two audits, conformal reliability of the next interval or frequency calibration of a closed window, and the two can disagree. For independent scores, restarting the empirical quantile at the boundary attains the minimax reliability rate, of order one over the square root of the post-break count. Restarting an adaptive recursion attains a frequency gap of order one over that count on bounded scores when its floor adds a bounded total. That gap bounds nothing about the next interval. After a large synthetic break, the restarted quantile's coverage error at the 50th forecast is 0.035, against 0.052 for the best of five adaptive baselines. The restarted recursion has the smallest frequency gap over the first 50 forecasts after every synthetic break. Real breaks are mostly smaller. The restarted quantile beats two baselines after an electricity market reform and trails all five after monetary policy decisions. Of nine announced breaks screened, only one is large enough for a detectable gain, and there every method misses nominal coverage by more than 0.1.
ER -