Should Legal Empiricists Go Bayesian?

Abstract:     
Bayesian empirical approaches appear frequently in fields such as engineering, computer science, political science and medicine, but almost never in law. This article illustrates how such approaches might be very useful in empirical legal studies. In particular, Bayesian approaches enable a much more natural connection between the normative or positive issues that typically motivate such studies and the empirical results.

Keywords: Bayesian analysis, right-to-carry, Bayesian model averaging, model, comparison, model selection, model specification, hypothesis testing, BIC, prior elicitation, prior sensitivity, conditional maximum, likelihood criterion, CML, maximum marginal likelihood criterion, MML, g-prior, natural conjugate

JEL Classifications: C10, C11, C12, C15, C23, C52, K14

Working Paper Series