Bayesian Benefits for Auditing

Statistical alternatives for improving audit quality and efficiency

About the Project

My name is Koen Derks. I am a PhD Candidate at Nyenrode Business University in collaboration with the University of Amsterdam. My project aims to develop novel Bayesian methods for statistical Auditing. It leans on two pillars of research:

1 – Improving availability of modern auditing techniques by programming these statistical techniques into the free, user-friendly and open-source software package JASP for Audit (JfA).

2 – Developing Bayesian alternatives to current statistical methods that are specifically suited to perform audit work, by combining various sources of data such that the amount of data that is needed to get to the desired level of audit risk is minimized.

    JASP for Audit (JfA)

    JfA is an open-source and free-of-charge module for JASP built to support the statistical aspects of an audit. The program provides a point-and-click interface, designed with the auditor in mind.

  • Planning

    JASP for Audit allows the auditor to calculate the required sample size for their audit. Bayesian methods make it possible to apply a sequential sampling plan.

  • Sampling

    JfA supports the auditor by guiding them through the sampling and stratification process. Multiple statistical sampling techniques are supported.

  • Evaluation

    JfA applies the correct scientifically-backed statistical techniques to provide a statement about the fairness of the population, both in Classical and Bayesian manifestation.

  • Interpretation

    JfA offers dynamic explanatory text corresponding to the displayed output, facilitating the correct interpretation of statistical audit results. Upon completion of an audit, JfA produces a full report with these results.




Collaborators

  • Nyenrode Business University is the thriving force behind this project, providing me with many opportunities to develop and share my work while being among experienced people from the field.

  • The Psychological Methods department of the University of Amsterdam is invaluable in the realization of this project. Their knowledge of Bayesian statistics is the foundation of the developed techniques.

  • I share a close collaboration with the development team behind JASP, which brings exiting opportunities for developing and implementing my software.

  • PricewaterhouseCoopers plays an advisory role in the project, providing insights and feedback from the practical audit field.