Mayo Matthew S, Gajewski Byron J
Department of Preventive Medicine and Public Health, Medical Statistics and Research Design Unit, Kansas Cancer Institute, University of Kansas Medical Center, Kansas City, KS, USA.
Control Clin Trials. 2004 Apr;25(2):157-67. doi: 10.1016/j.cct.2003.11.006.
A number of researchers have discussed phase II clinical trials from a Bayesian perspective. A recent article by Tan and Machin focuses on sample size calculations, which they determine by specifying a diffuse prior distribution and then calculating a posterior probability that the true response will exceed a prespecified target. In this article, we extend these sample size calculations to include informative prior distributions using various strategies that allow researchers with both optimistic and pessimistic priors direct involvement in the sample size decision making. We select the informative priors via multiple methods determined by the mean, median or mode of the conjugate prior. These cases can result in varying sample sizes.
许多研究人员从贝叶斯的角度讨论了II期临床试验。谭和马钦最近的一篇文章重点关注样本量计算，他们通过指定一个扩散先验分布，然后计算真实反应超过预先设定目标的后验概率来确定样本量。在本文中，我们将这些样本量计算方法进行扩展，以纳入信息性先验分布，使用各种策略使具有乐观和悲观先验的研究人员能够直接参与样本量决策。我们通过共轭先验的均值、中位数或众数确定的多种方法选择信息性先验。这些情况可能会导致不同的样本量。
