JASP for Audit: Bayesian Tools for the Auditing Practice

Abstract

Statistical theory is fundamental to many auditing guidelines and procedures. In order to assist auditors with the required statistical analyses, and to advocate state-of-the-art Bayesian methods, we introduce JASP for Audit (JfA). JfA is easy-to-use, free-of-charge software that automatically follows the standard audit workflow, selects the appropriate statistical analysis, interprets the results, and produces a readable report. This approach reduces the potential for statistical errors and therefore increases audit quality. Next to the frequentist methods that currently dominate audit practice, JfA incorporates Bayesian counterparts of these methods that come with several advantages. For example, Bayesian statistics allows incorporation of expert knowledge directly into the statistical analyses, allowing for a decrease in sample size, and an increase in efficiency. In sum, JfA is designed with the auditor in mind, it guides the auditor through the statistical aspects of an audit, and therefore has the potential to increase audit efficiency and quality.

Een Bayesiaanse blik op gestratificeerde steekproeven heeft voordelen voor de auditor

“…Onze hoop is dat, door deze methoden uiteindelijk beschikbaar te maken in de open-source software JASP for Audit (Derks et al 2021b), auditors meer gebruik zullen maken van Bayesiaanse hiërarchische modellen om steekproeven te evalueren.…”

Priors in a Bayesian audit: How integration of existing information into the prior distribution can improve audit transparency and efficiency

“…Finally, we have attempted to eliminate the difficulty of auditors determining the proper distributional family, specifying the prior, conducting sensitivity analyses, and documenting these choices, by implementing these methods in the R package "jfa" (Derks 2021) and JASP for Audit, a module for the freely available and open-source statistical software program JASP that is designed to facilitate Bayesian statistical auditing (Derks, de Swart, Wagenmakers, Wille, and Wetzels 2019). We hope that, by implementing these methods in a digital environment for which no extensive programming knowledge is required, they can be of aid to researchers and auditors interested in using Bayesian methods in audit sampling.…”

Priors in a Bayesian Audit: How Integration of Existing Information into the Prior Distribution Can Improve Audit Transparency and Efficiency

“…We have attempted to eliminate the difficulty of auditors determining the proper distributional family, specifying the prior, conducting sensitivity analyses, and documenting these choices, by implementing these methods in the open-source software JASP for Audit (JfA), a module for the freely available and open-source statistical software program JASP that is designed to facilitate Bayesian statistical auditing (Derks, de Swart, Wagenmakers, Wille, and Wetzels 2019). We hope that, by implementing these methods in a digital environment for which no programming knowledge is required, they can be of aid to researchers and auditors interested in using Bayesian methods to evaluate audit samples.…”