发表机构
University of Amsterdam; Nikhef(阿姆斯特丹大学; 荷兰国家亚原子物理研究所)
机构由 AI 辅助整理,请以论文原文为准。AI 中文总结
针对无密度、仅依赖模拟数据的序贯假设检验问题,构造 e-test 鞅,实现任意时刻有效的第一类错误控制、近似最优增长、几何衰减第二类错误及渐近功效为 1。
AI 中文摘要
对于给定的数据分布 $(X_t)_{t \in \mathbb{N}} \sim Q$ i.i.d.,我们研究假设检验问题:$H_0: Q = P_0$ 对比 $H_1: Q = P_1$,其中 $P_0$ 和 $P_1$ 是两个不同的模型概率分布。与标准设置(给定解析密度 $p_0$ 和 $p_1$)不同,这里我们考虑无密度设置,即我们只能访问 i.i.d. 模拟 $(Z^0_t)_{t \in \mathbb{N}} \sim P_0$ 和 $(Z^1_t)_{t \in \mathbb{N}} \sim P_1$。针对这种基于模拟的假设检验设置,我们构造了一个 e-test 鞅,从而得到一个具有任意时刻有效的第一类错误保证、近似增长最优性、几何衰减的第二类错误界限以及渐近功效为 1 的序贯检验。我们构造中使用的大部分成分都是众所周知概念的变体。本文的价值在于紧凑地呈现了一个针对无密度基于模拟的序贯假设检验情况的有效且任意时刻有效的解决方案。
英文摘要
For a given data distribution $(X_t)_{t \in \mathbb{N}} \sim Q$ i.i.d., we investigate the hypothesis testing problem: $H_0: Q = P_0$ vs. $H_1: Q = P_1$, for two different model probability distributions $P_0$ and $P_1$. In contrast to the standard setting, where analytic densities $p_0$ and $p_1$ are given, here, we consider the density-free setting, where we only have access to i.i.d. simulations $(Z^0_t)_{t \in \mathbb{N}} \sim P_0$ and $(Z^1_t)_{t \in \mathbb{N}} \sim P_1$. For this simulation-based hypothesis testing setting, we construct an e-test martingale, resulting in a sequential test with anytime-valid type-I error guarantees, approximate growth optimality, geometrically decaying type-II error bounds, and asymptotic power one. Most ingredients used in our constructions are variants of well known concepts. The value of this paper lies in the compact presentation of an effective, anytime-valid solution for the density-free simulation-based sequential hypothesis testing case.