Testing hypotheses under covariate-adaptive randomisation andadditive models

作     者：Ting Ye 

作者机构：Department of StatisticsUniversity of Wisconsin 1300 University Ave.MadisonWI 53706USA 

出 版 物：《Statistical Theory and Related Fields》 (统计理论及其应用（英文）)

年 卷 期：2018年第2卷第1期

页      面：96-101页

学科分类：07[理学] 0701[理学-数学] 070101[理学-基础数学] 

主　　题：Biased coin clinical trials robust test t-test type I error variance estimator 

摘      要：Covariate-adaptive randomisation has a long history of applications in clinical trials. Shao, Yu,and Zhong [(2010). A theory for testing hypotheses under covariate-adaptive ***, 97, 347–360] and Shao and Yu [(2013). Validity of tests under covariate-adaptivebiased coin randomization and generalized linear models. Biometrics, 69, 960–969] showed thatthe simple t-test is conservative under covariate-adaptive biased coin (CABC) randomisation interms of type I error, and proposed a valid test using the bootstrap. Under a general additivemodel with CABC randomisation, we construct a calibrated t-test that shares the same propertyas the bootstrap method in Shao et al. (2010), but do not need large computation required by thebootstrap method. Some simulation results are presented to show the finite sample performanceof the calibrated t-test.