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 current results. This temporal view of the data sequence is a convenience and does not play an active role in EB analysis. Much of the work in EB methods or so has been stimulated by H. Robbins in papers beginning with the reference Robbins where the terminology 'EB' was introduced. The various criteria for obtaining non-Bayes decision rules, and related special techniques, are well documented and will be assumed known. It will be seen that rules derived by the likelihood principle are prominent among the non-Bayes rules. In the EB approach the existence of a prior distribution is postulated, but it is taken to be susceptible to a frequency interpretation.
