Comparative Analysis of the Effectiveness of Three Models in Detecting Financial Statement Fraud

Abstrak

Financial statement transparency has become a critical issue in the post–Covid-19 period, particularly in the infrastructur, construction, property, and real estate sectors, which face financing pressures and complex revenue recognition. This study aims to compare the effectiveness of the Beneish M-Score, Dechow F-Score, and Altman Z-Score in detecting the risk of financial statement fraud (FSF). A quantitative approach is employed using classification analysis through a confusion matrix, complemented by ROC and AUC testing. The sample consists of 50 companies listed on the Indonesia Stock Exchange during the 2022–2024 period, with a total of 150 financial statement observations. The results indicate that the Altman Z-Score demonstrates the best performance, with an accuracy of 68%, an F1-Score of 62,5%, and an AUC of 0.742, followed by the Dechow F-Score and the Beneish M-Score. These findings suggest that financial distress is a key signal in identifying FSF. The study concludes that the Altman Z-Score is effective as an early warning system for financial statement fraud risk in the examined sectors.