Detection of Management Fraud: A Neural Network Approach

TLDR

Management fraud detection is a critical issue for auditors, with the Loebbecke and Willingham (1988) model and Bell et al.'s cascaded Logit approach serving as foundational methods. This study proposes an alternative detection method using Artificial Neural Networks (ANNs). The authors build a discriminator employing generalized adaptive neural network architectures (GANNA) and Adaptive Logic Network (ALN) designs. The resulting discriminant functions outperform Bell et al.'s cascaded Logit model in distinguishing fraudulent from non‑fraudulent firms and offer a concise set of diagnostic questions. © 1993.

Abstract

A bstract The detection of management fraud is an important issue facing the auditing profession. A major contributor to this issue is the Loebbecke and Willingham (1988) conceptual model for the detection of management fraud. A cascaded Logit approach using the Loebbecke and Willingham model was developed in Bell et al. (1993). The present study offers an alternative approach using Artificial Neural Networks (ANNs). This paper develops a successful discriminator of management fraud using both the generalized adaptive neural network architectures (GANNA) and the Adaptive Logic Network (ALN) approaches to designing neural networks. The discriminant functions can distinguish between fraudulent and non‐fraudulent companies with superior accuracy to the cascaded Logit results of Bell et al. (1993). Finally, the discriminant function provides a parsimonious set of questions useful for detecting management fraud.