{"id":"arxiv/2608.03889","url":"https://arxiv.org/abs/2608.03889","tags":["stat.ME","math.ST","stat.ML"],"score":0,"title":"Confidence Horizons","domain":"stat.ME","source":"arxiv","authors":["Chase Mathis","Ian Waudby-Smith"],"reviews":0,"subject":"stat.ME","summary":"Anytime-valid inference enables analysts to continuously monitor their data and stop experiments early. However, the majority of these methods incur a certain conservativeness by remaining valid on infinite time horizons. In practice, a bound on the horizon may be imposed due to budgetary, practical, or ethical constraints. In this paper, we ask the question: \"Is it possible to obtain sharper large-sample anytime-valid inference by forgoing validity beyond some finite time horizon?\". We provide a positive answer to this question by proposing a family of statistical objects that we call…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-08-04","reactions":{}}