Resumen de A Bayesian Model to Assess a Binary Measurement System When No Gold Standard System Is Available
Daniel P. Beavers, James D. Stamey, B.Neboyou Bekele
A repeated binary testing framework is a fallible means of determining whether or not units conform to quality standards. This article's Bayesian approach to the analysis of repeated binary testing data allows investigators to incorporate prior information on the conforming rate and misclassification parameters. A Markov chain Monte Carlo approach has been developed to represent posterior beta densities. The article also proposes methods for determining necessary sample sizes.
