Online monotone density estimation and log-optimal calibration

View PDF HTML (experimental)

Abstract:We study the problem of online monotone density estimation, where density estimators must be constructed in a predictable manner from sequentially observed data. We propose two online estimators: an online analogue of the classical Grenander estimator, and an expert aggregation estimator inspired by exponential weighting methods from the online learning literature. In the well-specified stochastic setting, where the underlying density is monotone, we show that the expected cumulative log-likelihood gap between the online estimators and the true density admits an $O(n^{1/3})$ bound. We further establish a $\sqrt{n\log{n}}$ pathwise regret bound for the expert aggregation estimator relative to the best offline monotone estimator chosen in hindsight, under minimal regularity assumptions on the observed sequence. As an application of independent interest, we show that the problem of constructing log-optimal p-to-e calibrators for sequential hypothesis testing can be formulated as an online monotone density estimation problem. We adapt the proposed estimators to build empirically adaptive p-to-e calibrators and establish their optimality. Numerical experiments illustrate the theoretical results.

Comments: 28 pages, 1 figure

Subjects:

Machine Learning (stat.ML); Machine Learning (cs.LG); Methodology (stat.ME)

Cite as: arXiv:2602.08927 [stat.ML]

(or arXiv:2602.08927v2 [stat.ML] for this version)

https://doi.org/10.48550/arXiv.2602.08927

arXiv-issued DOI via DataCite

Submission history

From: Rohan Hore [view email] [v1] Mon, 9 Feb 2026 17:29:39 UTC (101 KB) [v2] Mon, 30 Mar 2026 17:40:13 UTC (101 KB)