2022 Summer Institute Methods Lectures: Empirical Bayes Methods, Theory and Application

Large data sets that include observations on many workers at a given firm, multiple decisions by individual judges, auditors, and social insurance examiners, large numbers of students in the classrooms of specific teachers, and on many other settings with a similar structure, have generated substantial interest in estimating unit-specific parameters. Empirical Bayes methods make it possible to estimate these parameters, characterize their heterogeneity, and make decisions based on them. The 2022 Methods Lectures, presented by Jiayang Gu of the University of Toronto and Christopher Walters of the University of California, Berkeley, provide an introduction to the theory and application of these methods. A video recording of the two-part lecture series may be found above. The NBER also maintains an archive of Methods Lectures from the past 15 years.

Supported by the Alfred P. Sloan Foundation grant #G-2019-12304, the Lynde and Harry Bradley Foundation grant #20221137, and the National Science Foundation grant #1851757