统计与管理学院2015年学术报告第47期
【主 题】Assessing the similarity of two dose response curves
【报告人】Frank Bretz
Novartis
【时 间】 2015年11月05日（星期四）14:00－15:00
【地 点】 上海财经大学统计与管理学院大楼1208室
【语 言】 英文
【摘 要】This talk investigates the problem whether the difference between two parametric models describing the relation between a response variable and several covariates in two different groups is practically irrelevant, such that inference can be performed on the basis of the pooled sample. Statistical methodology is developed to demonstrate equivalence between the two regression curves by no more than a pre-specified relevance margin. More specifically, we derive a confidence interval for the maximum difference between the two given non-linear regression models over the entire covariate space of interest. If this confidence interval excludes the equivalence margins, similarity of dose response can be claimed. In addition, we consider bootstrap approaches as well as the delta method to derive confidence intervals for the difference between two same target doses. Again, if such a confidence interval excludes a pre-specified relevance margin, similarity in dose can be claimed.
【邀请人】尤进红
