Accounting for Uncertainty: An Application of Bayesian Methods to Accruals Models
47 Pages Posted: 16 Jul 2019
Date Written: July 9, 2019
We introduce a Bayesian approach for predicting “normal” accruals — a vital ingredient for measuring and identifying accrual-based earnings management. The literature’s standard approach takes a given model of normal accruals for granted, and neglects any prediction uncertainty. By contrast, our approach allows incorporating researchers’ uncertainty about the relevant models and parameters in the prediction of normal accruals. Our approach promises to increase power and reduce false positives in tests for opportunistic earnings management as a result of better predictions of normal accruals and more robust inferences. We advocate for greater use of Bayesian methods in accounting research, especially since they can now be easily implemented in popular statistical software packages.
Keywords: Accruals, Earnings Management, Prediction, Bayes
JEL Classification: C11, C53, M40
Suggested Citation: Suggested Citation