English

Significance improvement by randomized test in random sampling without replacement

Statistics Theory 2022-11-07 v1 Statistics Theory

Abstract

This paper studies one-sided hypothesis testing under random sampling without replacement. That is, when n+1n+1 binary random variables X1,,Xn+1X_1,\ldots, X_{n+1} are subject to a permutation invariant distribution and nn binary random variables X1,,XnX_1,\ldots, X_{n} are observed, we have proposed randomized tests with a randomization parameter for the upper confidence limit of the expectation of the (n+1)(n+1)th random variable Xn+1X_{n+1} under a given significance level δ>0\delta>0. Our proposed randomized test significantly improves over deterministic test unlike random sampling with replacement.

Keywords

Cite

@article{arxiv.2211.02399,
  title  = {Significance improvement by randomized test in random sampling without replacement},
  author = {Zihao Li and Huangjun Zhu and Masahito Hayashi},
  journal= {arXiv preprint arXiv:2211.02399},
  year   = {2022}
}
R2 v1 2026-06-28T05:11:01.649Z