arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.15468math.STstat.TH
收藏

两种实验设计下 Sobol' 与 Cramér-von Mises 指数的渐近效率

Asymptotic efficiency for Sobol' and Cram{é}r-von Mises indices under two designs of experiments

Thierry Klein, Agnès Lagnoux, Paul Rochet, Thi Mong Ngoc Nguyen

AI总结:

本文针对两种实验设计(Pick-Freeze 方案与 Given-Data 设定)推导了 Cramér-von Mises 指数的渐近效率界,并讨论了达到最优界的估计方法,填补了该指数渐近效率研究的空白。

摘要:

多种指数旨在量化输入变量对响应(通常是复杂计算机代码或黑箱模型的输出)的影响。最常用的 Sobol' 指数通常从解释方差的角度衡量某些输入的影响。然而,某些情况可能需要对某些输入的影响进行更具针对性的分析。在没有先验信息的情况下,基于分布的度量显得很有吸引力。为此,文献中提出了所谓的 Cramér-von Mises 指数(及其推广),其定义为在输出分布上积分的超额概率,旨在反映对输出整体分布的影响,而不仅仅是对方差的影响。这些不同指数的推断一直是一个具有挑战性的课题,尤其是在存在许多输入的情况下。虽然已知若干 Sobol' 指数估计量在正则性条件下是最优的,但迄今为止,文献中尚未解决 Cramér-von Mises 指数的渐近效率问题。针对这些指数,我们在本文中推导了效率界,并讨论了实现这些最优界的已知方法。考虑两种估计情境:所谓的 Pick-Freeze 方案和 Given-Data 设定,在后一种设定中,估计由唯一的输入-输出样本产生。

英文摘要:

A variety of indices aim to quantify the impact of input variables on a response, typically the output from a complex computer code or black-box model. Most commonly used, the Sobol' index typically measures the influence of some inputs from an explained variance perspective. However, some situations may require a more targeted analysis of some inputs influence. With no prior information, distribution-based measures appear to be appealing. In this purpose, so-called Cram{é}r-von Mises indices (and their generalization) have been proposed in the literature, defined as an excess probability integrated over the output distribution that aim to reflect influence on the whole distribution of the output rather than on the variance solely. Inference of these various indices has remained a challenging topic especially in presence of many inputs. While several Sobol' indices estimators are known to be optimal under regularity conditions, the issue of asymptotic efficiency for Cram{é}r-von Mises indices has been unaddressed in the literature so far. For these indices, we derive in this paper the efficiency bounds and discuss the known methods to achieve such optimal bounds. Two estimation contexts are considered: the so-called Pick-Freeze scheme and the Given-Data setting, for which the estimation is produced from a unique input-output sample.

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。

↑