{"id":"arxiv/2603.06941","url":"https://arxiv.org/abs/2603.06941","tags":["math.ST","stat.ME"],"score":0,"title":"Demonstration Experiments","domain":"math.ST","source":"arxiv","authors":["Guido Imbens","Lorenzo Masoero","Alexander Rakhlin","Thomas S. Richardson","Suhas Vijaykumar"],"reviews":0,"subject":"math.ST","summary":"Adaptive experiments are used extensively in online platforms, healthcare and biotechnology, and the social sciences. Often, the primary goal is not to precisely estimate a treatment effect but to demonstrate that at least one candidate intervention yields a positive effect, for some subpopulation and on some measured outcome. We formalize this objective as testing the global null in a threshold bandit framework, and develop two inference procedures that are valid under general adaptive sampling: one that pools information across promising arms, and one based on time-uniform multiple testing…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-06","reactions":{}}