Paperlane

no. 16 · Tuesday 17 November 2026

Honest confidence sets under experiment-level failure

Yuxin Han, Lanjun Cheng and Fengshi Jing

math.ST · via stat.ME

Abstract

An experiment-level failure replaces the joint observation law while preserving the target. We study honest confidence sets when a declared upper bound on failure probability equals the allowed noncoverage probability. For a Uniform support endpoint and a Gaussian mean on fixed compact parameter intervals, every honest deterministic procedure has full-range worst-case expected diameter under arbitrary or absolutely continuous joint failures. Independent randomisation gives sharp minimax risks of $(2+o(1))\log N/N$ for the endpoint in $[1,2]$ and $(2σ+o(1))(\log N/N)^{1/2}$ for the mean with known variance $σ^2$. A failure distribution shared by many targets forces average clean-state noncoverage to decrease with expected diameter. One-sided local Kullback–Leibler bounds then yield lower bounds with logarithmic factors. The endpoint experiment also has a sharp deterministic frontier as the remaining error budget vanishes. Under exchangeable failures without ties, observation order recovers both randomised benchmark rates deterministically. In a reference-calibrated roster, we derive the randomised minimax reliability rate. A prespecified reference reserve of logarithmic size permits deterministic attainment of the same order on a subclass whose conditionally independent reserved responses have probabilities bounded away from zero and one in every state.

fields
Statistics Theory
pdf ↗ page arXiv:2611.20079 ↗