A Bayesian Audit Assurance Model Incorporating Monetary Unit Sampling

42 Pages Posted: 7 May 2024

See all articles by Trevor Stewart

Trevor Stewart

Rutgers, The State University of New Jersey

Daniel Sunderland

Northeastern University

Date Written: May 5, 2024


The process for obtaining reasonable audit assurance described by modern auditing standards—iteratively updating subjective professional judgement for new evidence—is essentially Bayesian. Recognizing it as such unlocks a box of tools and techniques for modeling audit assurance, designing targeted audit procedures, and updating the model for the results. We describe such a model and focus on its interrelationship with monetary unit sampling. The model arises naturally as a generalization of the familiar audit risk model, fully integrates sampling results into the overall assurance profile, has intuitive auditor appeal, and is technically easy to implement for field deployment. We show how two audit judgments serve to uniquely identify an appropriate prior, how a third fixes the target posterior, and how sample sizes are derived via Bayes’ rule. We introduce the concept of the Stringer posterior, a Bayesian adaptation of the Stringer bound familiar from the monetary unit sampling literature.

Keywords: Bayesian audit assurance, monetary unit sampling, audit risk model, prior assurance, assurance profile, Stringer bound, Stringer posterior, Poisson distribution, gamma distribution

JEL Classification: C11, M42

Suggested Citation

Stewart, Trevor and Sunderland, Daniel, A Bayesian Audit Assurance Model Incorporating Monetary Unit Sampling (May 5, 2024). Available at SSRN: https://ssrn.com/abstract=4817734 or http://dx.doi.org/10.2139/ssrn.4817734

Trevor Stewart (Contact Author)

Rutgers, The State University of New Jersey ( email )

One Washington Park, Room 919
Newark, NJ 07102
United States

Daniel Sunderland

Northeastern University ( email )

220 B RP
Boston, MA 02115
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics