摘要
在无金标准二重抽样设计下,本文从疾病流行率的置信区间的角度考虑了临床试验所需样本量的确定。分别在条件独立和非独立模型下,基于疾病流行率的Wald置信区间、逆双曲正切变换置信区间给出了在给定置信水平下区间宽度控制在指定范围内的样本量估计方法;此外,在条件独立模型下,基于贝叶斯可信区间也给出了区间宽度控制下的样本量估计方法。模拟研究考察了在估计的样本量下各种置信区间的统计性质,通过实例分析数据说明了本文提出方法的实际应用。
Under the double-sampling design without gold standard,this paper investigates the determination of sample sizes for disease prevalence studies from the perspective of the confidence interval.Sample size determination methods are developed to control the width of a confidence interval with a pre-specified confidence level.We provide sample size determination methods based on Wald confidence interval(CI)and the inverse hyperbolic tangent transformation CI of the disease prevalence under the conditional independent model and dependent model,and under the conditional independent model,the sample size estimator based on a Bayesian credible interval is also developed.Simulation studies have been conducted to demonstrate the performance of various CIs with the proposed sample sizes.A realdata example is used to illustrate the applicability of the proposed methods.
作者
邱世芳
瞿颖秋
张晓良
曾莎
QIU Shi-fang;QU Ying-qiu;ZHANG Xiao-liang;ZENG Sha(College of Science,Chongqing University of Technology,Chongqing 400054,China)
出处
《数理统计与管理》
CSSCI
北大核心
2024年第2期249-262,共14页
Journal of Applied Statistics and Management
基金
国家自然科学基金项目(11871124,11471060)
重庆市基础研究与前沿探索项目(cstc2018jcyjAX0241)。
关键词
无金标准
二重抽样
区间宽度
样本量
贝叶斯可信区间
imperfect gold standard
double-sampling
interval width
sample size
Bayesian credible interval
