Fighting Financial Statement Fraud Through Predictive Analytics

• In Event: Identifying and Regulating White-Collar and Corporate Crime

Abstract

Corporate crimes fall into several categories out of which Financial Statement Fraud (FSF) draws a great deal of attention since its costs and losses are greatly under-reported causing serious consequences to the welfare of the community. FSF is defined as the calculated misrepresentation of the financial statement information that is publicly disclosed by companies in order to mislead stakeholders regarding the firm's true financial position.

Standard auditing procedures are insufficient to identify fraudulent reporting since most managers comprehend the limitations of audits, hence the need for unbiased analytic tools that help auditors to differentiate between fraud and non-fraud cases. Although several statistical models have been developed for this end, these can be improved in terms of data and accuracy. Accordingly, the main objective of this study is to develop data-informed methods to be used in combination with the experience and instinct of experts to lead to better decision-making in terms of regulatory policy and strategy.

Compared to previous studies, a significantly larger sample is used in this article that includes all public releases issued by the SEC between 1990 and 2012. Generally, results suggest that there is potential in detecting FSF through predictive analytics and publicly available financial statement information.

Authors

• Maria Jofre, The University of Sydney

• Richard Gerlach, The University of Sydney