Bayesian methods to estimate the accuracy of a binary measurement system.
Access rightsNo access - Contact firstname.lastname@example.org
Eschmann, Mark, 1984-
MetadataShow full item record
Binary Measurement Systems (BMS) are frequently used in such applications as quality control. They are important enough that their operating characteristics, the repeatability and reproducibility are important because of the clues that provide engineers in improving the BMS. We have developed a Bayesian single inspector model that incorporates baseline information. Simulation studies show that this can cut credible set width of the misclassification parameters in half. We have also developed Bayesian fixed effects and random effects models for BMS's with multiple inspectors. Simulation studies that demonstrated acceptable performance characteristics. Several applications are also worked out included sample size determination for the fixed effects model, and superior subset selection with the random effects model.