小样本内部预试验自适应设计中Ⅰ型错误和检验效能的控制及模拟

摘要

目的 通过盲态下内部预试验IPS( internal pilot study)样本量调整的模拟分析,探索小样本下有效控制Ⅰ型错误、保证检验效能的合理统计方法.方法 利用蒙特-卡罗( Monte Carlo)模拟不断改变两总体间的均值差,固定方差,在盲态下采用随机化检验进行统计分析,计算Ⅰ型错误和检验效能随均值差的变化情况.结果 盲态下Ⅰ型错误未超过检验水准,检验效能略有降低.结论 盲态小样本下调整样本量时,采用随机化检验可有效控制Ⅰ型错误.%Objective To determine the appropriate methods to control the inflation of type I error and the lowering of power on the condition of small sample size through simulations of analysis for blinding Internal Pilot Study in clinical trials. Methods Changed the mean difference of the two populations and fixed the variance,used randomization test to analyze the final data and calculate type I error and the power with the changing of the mean difference. Results Type I error rate was not beyond the significance level and the powers were contained. Conclusion Randomization test could control the inflation of type I error effectively on the condition of blinding sample, size adjustment with small sample size.