{"id":"arxiv/2611.20079","url":"https://arxiv.org/abs/2611.20079","tags":["stat.ME","math.ST"],"score":0,"title":"Honest confidence sets under experiment-level failure","domain":"stat.ME","source":"arxiv","authors":["Yuxin Han","Lanjun Cheng","Fengshi Jing"],"reviews":0,"subject":"stat.ME","summary":"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…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-11-15","reactions":{}}