arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1502.07864stat.CO
收藏

多假设检验的蒙特卡罗模拟最优分配

Optimal allocation of Monte Carlo simulations to multiple hypothesis tests

Georg Hahn

摘要:

在实践中,多个假设检验通常使用自举或排列检验获得的p值估计值进行,因为所有假设的分析p值通常是未知的。本文考虑分配预先指定总数的蒙特卡罗模拟$K \in \mathbb{N}$(即,排列或从自举分布中抽取)到给定数目的$m \in \mathbb{N}$假设,以便以最佳方式近似它们的p值$p \in [0,1]^m$,在这种意义上,分配使错误分类的假设的总预期数目最小化。如果用近似p值获得的单个假设的决策与其p值通过分析已知时获得的决策不同,则会发生错误分类。本文的贡献有三个方面:在已知$p$和$K \in \mathbb{R}$的假设下,并使用二项分布的正态近似,当使用Bonferroni校正校正多重性时,计算有或没有伪计数的p值估计值时,推导出$K$模拟到$m$假设的最佳实值分配。在前一种情况下产生的计算微妙之处将被讨论。第二,在基于模拟退火算法的帮助下,经验证据表明,最优整数分配可能与最优实值分配具有相同的形式,并且两者似乎是一致的。第三,对模拟和真实数据的实证研究表明,最近提出的基于Thompson采样的采样算法,当p值未知时,可以近似地模拟最优(实值)分配,从而在运行时进行估计。

英文摘要:

Multiple hypothesis tests are often carried out in practice using p-value estimates obtained with bootstrap or permutation tests since the analytical p-values underlying all hypotheses are usually unknown. This article considers the allocation of a pre-specified total number of Monte Carlo simulations $K \in \mathbb{N}$ (i.e., permutations or draws from a bootstrap distribution) to a given number of $m \in \mathbb{N}$ hypotheses in order to approximate their p-values $p \in [0,1]^m$ in an optimal way, in the sense that the allocation minimises the total expected number of misclassified hypotheses. A misclassification occurs if a decision on a single hypothesis, obtained with an approximated p-value, differs from the one obtained if its p-value was known analytically. The contribution of this article is threefold: Under the assumption that $p$ is known and $K \in \mathbb{R}$, and using a normal approximation of the Binomial distribution, the optimal real-valued allocation of $K$ simulations to $m$ hypotheses is derived when correcting for multiplicity with the Bonferroni correction, both when computing the p-value estimates with or without a pseudo-count. Computational subtleties arising in the former case will be discussed. Second, with the help of an algorithm based on simulated annealing, empirical evidence is given that the optimal integer allocation is likely of the same form as the optimal real-valued allocation, and that both seem to coincide asympotically. Third, an empirical study on simulated and real data demonstrates that a recently proposed sampling algorithm based on Thompson sampling asympotically mimics the optimal (real-valued) allocation when the p-values are unknown and thus estimated at runtime.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑