カワサキ　ヨウヘイ   KAWASAKI Yohei   川﨑　洋平    所属   埼玉医科大学  医学部 研究統括センター    職種   教授
論文種別	学術雑誌(原著)
言語種別	英語
査読の有無	査読あり
表題	A Bayesian inference of P(pi1 > pi2) for two proportions.
掲載誌名	正式名：Journal of biopharmaceutical statistics ISSNコード：15205711/10543406
掲載区分	国外
巻・号・頁	22,425-37頁
著者・共著者	Kawasaki Y, Miyaoka E
発行年月	2012
概要	The statistical inference concerning the difference between two independent binominal proportions is often discussed in medical and statistical literature. However, such discussions are often based on the frequentist viewpoint rather than the Bayesian viewpoint. In this article, we propose an index theta =P(pi(1, post ) > pi(2, post )), where pi(1, post ) and pi(2, post ) denote binominal proportions following posterior density. We provide approximate and exact expressions for theta by using the beta prior. We also present the results of actual clinical trials to show the utility of theta. Our findings suggest that theta can potentially provide useful information in a clinical trial.