ABSTRACT
This chapter summaries some of the applications of Empirical Bayes (EB) methods that have been published. Data from non-trivial practical applications seldom follow the somewhat idealized patterns assumed in the development of techniques. The EB sampling schemes which have been studied in much detail in previous chapters, and in a large body of other EB literature, are relatively simple, and few actual data sets follow those patterns exactly. If a study is aimed at searching for or quantifying differences between groups of subjects it is not clear that EB methods are relevant. The basic data used in the EB analysis comprises two least squares regression coefficients together with an estimated covariance matrix for each law school. The EB estimates, actually referred to as James-Stein estimates by R. E. Fay and R. A. Herriot, exhibit superior performance relative to two other, more common, estimates.
