Financial Statement Fraud Detection: An Analysis of Statistical and Machine Learning Algorithms

TLDR

The study compares the performance of six statistical and machine learning models for detecting financial statement fraud under varying misclassification costs and fraud‑to‑nonfraud ratios. The authors evaluate these models by measuring their detection accuracy across different cost and ratio scenarios. Logistic regression and support vector machines outperform the other models, with only six predictors—auditor turnover, total discretionary accruals, Big 4 auditor, accounts receivable, meeting or beating analyst forecasts, and unexpected employee productivity—consistently selected, offering insights that can improve fraud risk models for practitioners and regulators. Data on fraud companies are available from the author upon request, and all other data sources are described in the text.

Abstract

SUMMARY This study compares the performance of six popular statistical and machine learning models in detecting financial statement fraud under different assumptions of misclassification costs and ratios of fraud firms to nonfraud firms. The results show, somewhat surprisingly, that logistic regression and support vector machines perform well relative to an artificial neural network, bagging, C4.5, and stacking. The results also reveal some diversity in predictors used across the classification algorithms. Out of 42 predictors examined, only six are consistently selected and used by different classification algorithms: auditor turnover, total discretionary accruals, Big 4 auditor, accounts receivable, meeting or beating analyst forecasts, and unexpected employee productivity. These findings extend financial statement fraud research and can be used by practitioners and regulators to improve fraud risk models. Data Availability: A list of fraud companies used in this study is available from the author upon request. All other data sources are described in the text.