{"id":"arxiv/2606.05072","url":"https://arxiv.org/abs/2606.05072","tags":["math.ST"],"score":0,"title":"Adaptive Sequential Change Detection using Mixtures of Predictive Distributions","domain":"math.ST","source":"arxiv","authors":["Topi Halme","H. Vincent Poor","Visa Koivunen"],"reviews":0,"subject":"math.ST","summary":"This paper studies the problem of detecting a change in the distribution of a sequence of independent observations when the post-change distribution is unknown. We propose a novel change detection algorithm, termed Predictive-Mixture CuSum (PM-CuSum), which combines predictive distributions constructed from sliding windows of different lengths within a CuSum recursion. The predictive distributions are aggregated using adaptive weights based on their recent predictive performance. We show that PM-CuSum achieves first-order asymptotic optimality under mild conditions, and that its asymptotic…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-06-03","reactions":{}}