摘要：Bayesian latent-class models have b een widely applied for assessing the performanceof diagnostic tests in the absence of a gold standard. We provide a short discussion onidentifiability issues appearing under the absence of a gold standard, and construct anextension of the well-known Hui–Walter stratification mo del which allows for stratum-dependent parameters. We illustrate our approach using a Chagas disease case studyon blood donors from Brazil
关键词：absence of a gold standard; diagnostic test; identifiability; sample size; stratification
