ABSTRACT
Empirical Bayes (EB) and related techniques come into play when data are generated by repeated execution of the same type of random experiment. EB methods provide a way in which such historical data can be used in the assessment of the results. This temporal view of the data sequence is a convenience and does not play an active role in EB analysis. Applications of the EB methods described in this monograph are envisaged as occurring in repetitive experimentation with parameters varying from experiment to experiment. The EB approach can be essentially non-Bayesian in the sense of not involving subjective probabilities. The general idea of the developments is that the EB technique is an attractive compromise between the conventional non-Bayes approach and the fully specified Bayesian approach for the analysis of historical data arising in a sampling scheme.
