A Bayesian Approach to Event Studies for Securities Litigation

摘要

The paper by Gelbach and Hawkins (2020) is a superb overview of a Bayesian treatment of event studies with which we are in complete agreement. Event studies used in litigation typically involve the use of frequentist statistical inference via the classical approach of null-hypothesis significance testing. Gelbach and Hawkins (2020) offer a well-reasoned Bayesian alternative and give an illustration using a case study that comes from litigation involving events that occurred in 2004 concerning Apollo Group, Inc., the parent company of the University of Phoenix. We give some general comments about the Bayes approaches as well as two reanalyses of the Apollo Group data.