Fidler V
Biometrics. 1984 Dec;40(4):1063-70.
A mixed model is proposed for the two-period change-over clinical trial with binary responses. Tests for the treatment effect are derived under several versions of this model. This approach reveals the relative merits of tests previously described in the literature and leads to a new test for the carry-over effect.
针对具有二元响应的两阶段交叉临床试验,提出了一种混合模型。在该模型的几个版本下推导了治疗效果的检验。这种方法揭示了文献中先前描述的检验的相对优点,并得出了一种新的残留效应检验方法。