Designing a Tax Fraud Detection Model Using Financial Statements
Keywords:
Tax fraud, F-Score, Z-Score, Logistic regression, Manufacturing companies, Financial indicatorsAbstract
The objective of this study is to design and test a predictive model for identifying tax fraud in manufacturing companies listed on the Tehran Stock Exchange. The study used data from 100 manufacturing firms covering the years 2015–2024. Key predictors included F-Score for financial statement manipulation, Z-Score for bankruptcy risk, leverage, liquidity, profitability, asset composition, and new share issuance. A multivariate logistic regression model was developed and evaluated using collinearity diagnostics, logit linearity tests, model fit indices, Pseudo R² values, confusion matrix analysis, and ROC curve assessment. The findings revealed that F-Score and leverage exert strong positive effects on the probability of tax fraud, while Z-Score, liquidity, and profitability reduce this likelihood. Asset complexity and new share issuance increase fraud risk. The model demonstrated an accuracy of 86.5%, sensitivity of 72.5%, specificity of 90%, and an AUC of 0.88. Interaction analysis indicated that the joint presence of financial pressure and financial manipulation substantially amplifies fraud risk. The proposed model integrates indicators of financial pressure, financial manipulation, and opportunity into a comprehensive analytical framework, enabling accurate prediction of tax fraud risk. It offers practical value as a decision-support tool for tax risk management and audit prioritization.
Downloads
References
Aghghaleh, S., Mohammadi, M., & Shabanpour, H. (2016). Predicting financial statement fraud using Dechow F-Score model. Journal of Financial Crime, 23(4), 1045-1062.
Al-Ali, A., Khedr, A. M., El-Bannany, M., & Kanakkayil, S. (2023). A powerful predicting model for financial statement fraud based on optimized XGBoost ensemble learning technique. Applied Sciences, 13(4), 2272. https://doi.org/10.3390/app13042272
Al-Hiyari, F., & et al. (2022). Detection of fraudulent financial papers by picking a collection of characteristics using optimization algorithms and classification techniques based on squirrels. https://arxiv.org
Ali, A., Abd Razak, S., Othman, S. H., Eisa, T. A. E., Al Dhaqm, A., Nasser, M., Elhassan, T., Elshafie, H., & Saif, A. (2022). Financial fraud detection based on machine learning: A systematic literature review. Applied Sciences, 12(19), 9637. https://doi.org/10.3390/app12199637
Amin, M. (2018). Factors affecting financial statement fraud: Evidence from corporate reports. International Journal of Accounting Studies, 5(2), 55-72.
Bagherian, R., Kasgari, M., & Amiri, M. (2022). Financial and non-financial determinants of fraud in Iranian public companies: The role of machine learning. Journal of Accounting & Organizational Change, 18(2), 311-330.
Dechow, P. M., Ge, W., & Schrand, C. (2011). Understanding earnings quality: A review of the proxies, their determinants and their consequences. Journal of Accounting and Economics, 50(2-3), 344-401. https://doi.org/10.1016/j.jacceco.2010.09.001
Farber, D. B. (2005). Restoring trust after fraud: Does corporate governance matter? The Accounting Review, 80(2), 539-561. https://doi.org/10.2308/accr.2005.80.2.539
Ginanjar, A., & Wibowo, A. (2025). Data Mining to Detect Fraud Patterns in a Taxpayer's Financial Statement. Scientax, 6, 135-150. https://doi.org/10.52869/st.v6i2.571
Hossain, M. Z., Raja, M. R., & Hasan, L. (2024). Developing predictive models for detecting financial statement fraud: A machine learning approach. European Journal of Theoretical and Applied Sciences, 2(6), Article 22. https://doi.org/10.59324/ejtas.2024.2(6).22
Kamrani, H., & Abedini, B. (2023). Developing a Financial Statement Fraud Detection Model Using Artificial Neural Network and Support Vector Machine Methods in Companies Listed on the Tehran Stock Exchange. Scientific Research Journal of Accounting and Financial Management.
Kou, G., Peng, Y., & Wang, G. (2014). Evaluation of classification algorithms using MCDM and rank correlation. International Journal of Information Technology & Decision Making, 13(01), 31-52.
Kumar, S. (2021). Financial ratios and fraud detection in emerging markets. Journal of Financial Crime, 28(3), 765-783.
Lundberg, S. M., & Lee, S. I. (2017). A unified approach to interpreting model predictions.
Majdam, A., Charmahalizadeh, S., Kohandel, M., & Nouralizadeh, N. (2023). Presenting a Model for Identifying Financial Statement Fraud Based on Psychological Components Through Grounded Theory and Meta-Synthesis. Iranian Journal of Accounting and Economic Sciences.
Mehr News, A. (2019). What is the Real Figure of Tax Evasion in Iran's Economy? https://www.mehrnews.com/news/4769064/
Nguyen, T. (2024). Identifying deceptive financial reporting via analysis of financial statements: Case in Vietnam. Public SCNCHub. https://public.scnchub.com
Piri, P., & Kazemi, T. (2022). Predicting Financial Reporting Fraud Schemes Using a Multi-Class Machine Learning Approach. Journal of Accounting and Financial Management Research.
Price Iii, R., Bansal, S., & Sharma, P. (2011). Empirical evaluation of Dechow F-Score model in fraud detection. Journal of Forensic Accounting Research, 12(1), 77-101.
Rahimikia, A., Mohammadi, A., Rahmani, M., & Ghazanfari, M. (2017). Detection of corporate tax evasion in Iran using machine learning methods. International Journal of Financial Studies, 5(4), 35.
Reeb, D., & et al. (2025). Using machine learning to detect financial statement fraud: A cross-country analysis applied to Wirecard AG. https://www.mdpi.com
Rezaei, M., Nazemi Ardakani, M., & Nasser Sadrabadi, A. (2021). Detection of Financial Statement Fraud Considering the Financial Statement Audit Report. Iranian Journal of Accounting and Financial Research.
Summers, S. L., & Sweeney, A. P. (1998). Fraudulent financial reporting: 1987-1997: An analysis of US public companies. Accounting Horizons, 12(4), 107-125.
Zhang, X., Li, Y., & Chen, J. (2023). Detecting financial statement fraud using dynamic ensemble machine learning. ScienceDirect. https://www.sciencedirect.com
Downloads
Published
Submitted
Revised
Accepted
Issue
Section
License
Copyright (c) 2025 Tohid Seyfollahzadeh Sarai (Author); Amin Najafgholizadeh (Corresponding author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
