Can Governance and Forensic Accounting Metrics Predict Stock Returns?

Abstract

Arguably, governance and forensic accounting metrics should be predictors of future stock returns. Do empirical tests confirm this view? This article describes the logic and analytics used to calculate a metric called the Forensic Alpha Model (FAM), which was then used to test the hypothesis that it could predict future stock returns. The results of these tests, conducted with out-of-sample data, confirm that forensic accounting and governance metrics did indeed predict future stock returns. Implementing the FAM requires enormous data collection, detailed peering, industry normalizations, forensic accounting and governance taxonomies, sophisticated measures of association and interactions, rigorous testing, and advanced supervised machine learning.

Keywords: Forensic Accounting, Fraud, Governance, Neural Networks, Pension Fund, Supervised Learning