Baker S G
Division of Cancer Prevention and Control, National Cancer Institute, Bethesda, Maryland 20892-7354, USA.
Biometrics. 1996 Mar;52(1):362-9.
In analyzing partially observed categorical case-control data, Williamson and Haber (1994, Biometrics 50, 194-203) assumed ignorable nonresponse. We reanalyzed the data allowing for nonignorable nonresponse while explicitly incorporating case-control sampling.
在分析部分观测的分类病例对照数据时,威廉姆森和哈伯(1994年,《生物统计学》第50卷,第194 - 203页)假定存在可忽略的无应答情况。我们重新分析了这些数据,考虑到不可忽略的无应答情况,同时明确纳入病例对照抽样。