{"id":"arxiv/2610.24668","url":"https://arxiv.org/abs/2610.24668","tags":["math.PR"],"score":0,"title":"On the Proportion of Time that a Conventional Confidence Interval Covers","domain":"math.PR","source":"arxiv","authors":["Monte Fischer","Peter Glynn"],"reviews":0,"subject":"math.PR","summary":"The concept of confidence sequences, or `anytime confidence intervals,' has attracted significant interest in recent years. This motivates the question of how well conventional confidence intervals behave when similarly evaluated. In particular, we study the probability distribution of the proportion of time a conventional confidence interval procedure covers the population mean, when the total time $n$ spent sampling is large. We establish an invariance principle for the limiting distribution as $n$ tends to infinity and show that the limit random variable $Z(1)$ can be expressed as a…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-17","reactions":{}}