Leveraging the Fisher Randomization Test using Confidence Distributions: Inference, Combination and Fusion Learning
Leveraging the Fisher Randomization Test using Confidence Distributions: Inference, Combination and Fusion Learning
复制标题
利用置信分布的 Fisher 随机化检验：推理、组合和融合学习
DOI：
10.1111/rssb.12429
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Liu, Regina Y.
中科院分区：
文献类型：
--
作者：
Luo, Xiaokang;Dasgupta, Tirthankar;Xie, Minge;Liu, Regina Y.
Fisher随机化检验（FRT）的灵活性和广泛适用性使其成为评估现代随机化实验干预措施因果效应的有吸引力的工具，这些实验的规模和复杂性都在增加。本文通过建立FRT与置信分布的联系，为FRT提供了一个理论推导框架。这样的连接导致的发展的（i）一个明确的程序反演的FRT生成置信区间的保证覆盖率，（ii）新的见解的影响的Monte Carlo样本的大小估计的ap值曲线和（iii）通用和特定的方法，联合收割机FRT从多个独立的实验与理论保证。我们的发展涉及有限的样本设置，但有直接扩展到大样本。仿真和一个案例证明了这些新的发展的好处。
The flexibility and wide applicability of the Fisher randomization test (FRT) make it an attractive tool for assessment of causal effects of interventions from modern-day randomized experiments that are increasing in size and complexity. This paper provides a theoretical inferential framework for FRT by establishing its connection with confidence distributions. Such a connection leads to development’s of (i) an unambiguous procedure for inversion of FRTs to generate confidence intervals with guaranteed coverage, (ii) new insights on the effect of size of the Monte Carlo sample on the estimation of ap-value curve and (iii) generic and specific methods to combine FRTs from multiple independent experiments with theoretical guarantees. Our developments pertain to finite sample settings but have direct extensions to large samples. Simulations and a case example demonstrate the benefit of these new developments.