Abstract
This paper examines a Bayesian rule recently suggested for the classification of inspectors in sensory tests. The rule is presented and shown to be equivalent to a Bayesian hypothesis testing procedure, which aids in the interpretation of the rule. The rule is then improved upon by making fuller use of available sample data. Finally, an extension to the case of unknown error probabilities within inspector classes is briefly described.
Additional information
Notes on contributors
Christopher B. Barry
Dr. Barry is President of Micro Data Base Systems, Inc. He was an Associate Professor of Finance in the Graduate School of Business at the University of Texas at Austin when this article was written.