Corporate Financial Fraud—Assessing a Multi-Disciplinary Risk Factors Framework with Machine Learning

• In Event: Exploring the Role of Risk in White-Collar/Corporate Crime

Abstract

Corporate Financial Reporting and Statements Fraud (FSRF) is one of the costliest forms of white-collar crime (ACFE, 2018) with far-reaching consequences. While corporate crime theories have presented us with various hypothesized antecedents, limited effort has been made to empirically examine risk factors of FSRF. Using manually coded data from official SEC enforcement releases, the current study examines a set of multi-disciplinary risk factors of FSRF pertaining to linguistics cues, organizational structure and financial metrics. These risk factors are unified and explained under a theoretical framework and are analyzed using machine learning algorithms to assess their aggregate predictive power.

Author

• Fiona Chan, Michigan State University