Volume 33, Issue 68 (2026)
J Tax Res 2026, 33(68): 125-170 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
DABBAGH R, ezzati shourghouli A, Shabanpoor R. Comparison of the Performance of Machine Learning Algorithms Based on Artificial Intelligence with the State Space Model in Forecasting Tax Revenues of West Azerbaijan Province. J Tax Res 2026; 33 (68) :125-170
URL: http://taxjournal.ir/article-1-2434-en.html
URL: http://taxjournal.ir/article-1-2434-en.html
1- Afagh Higher Education , R.Dabbagh@uut.ac.ir
2- Urmia University of Technology
2- Urmia University of Technology
Abstract: (235 Views)
Tax revenues are among the most significant sources of government income, comprising a substantial portion of the government budget. These revenues are essential for funding public services, implementing infrastructure projects, and covering administrative and operational expenses. Accurate tax revenue forecasting is a critical tool for economic planning and decision-making. Reliable forecasts enable governments to plan budgets effectively, formulate optimal tax policies, and enhance economic management by reducing uncertainties. This study uses monthly tax revenue data from West Azerbaijan Province spanning April 2005 to June 2024 to forecast tax revenues using machine learning algorithms, including Random Forest, Support Vector Regression, Lasso Regression, Deep Learning, and the State Space Model. The findings indicate that among the evaluated algorithms—Random Forest, Support Vector Regression, Lasso Regression, and Deep Learning—the Deep Learning model achieves the highest accuracy in both in-sample and out-of-sample forecasts based on RMSE (Root Mean Square Error) and MAPE (Mean Absolute Percentage Error) metrics. However, the State Space Model emerges as a robust tool for modeling and forecasting financial and economic variables, particularly those exhibiting seasonal, monthly, or daily patterns. Its ability to capture unobservable components, irregularities, and dynamic trends provides a notable advantage in analyzing complex economic data. A comparative analysis of the State Space Model as a robust econometric approach and Deep Learning as a cutting-edge machine learning method demonstrates that the State Space Model outperforms Deep Learning in forecasting tax revenues for West Azerbaijan Province.
Keywords: West Azerbaijan Province, Artificial Intelligence (AI), Deep Learning, Tax Revenue Forecasting
Type of Study: Research |
Subject:
Economic
Received: 2024/10/25 | Accepted: 2025/03/18 | Published: 2026/03/1
Received: 2024/10/25 | Accepted: 2025/03/18 | Published: 2026/03/1
References
1. دباغ، رحیم، فرزان، محمد و محسنی، بهزاد. (1404). شناسایی چالشهای فناوریهای صنعت 4.0 برای عملیات پایدار در صنایع کوچک و متوسط. پژوهش در مدیریت تولید و عملیات، 16(1)، 1-24.
2. دباغ، رحیم، گلمرادی، باقری. (1398). بررسی عملکرد مالی بانکهای اسلامی و غیراسلامی در کشورهای منتخب با استفاده از مدل CAMEL. فصلنامه مطالعاتی در مدیریت بانکی و بانکداری اسلامی، 4(پاییز)، 85-114.
3. زارعی، پریسا، جلائی، سید عبدالمجید و صادقی، زین العابدین (۱۳۹۸). شبیه سازی و پیش بینی اثر مالیات سبز بر مصرف و شدت انرژی در ایران با استفاده از الگوریتم ژنتیک. پژوهشنامه مالیات، ۲۷(۴۲)، ۱۰۳-۱۲۵
4. سعادت مهر، مسعود. (1400). بررسی تأثیر متغیرهای کلان اقتصاد بر درآمدهای مالیاتی دولت در ایران. دوفصلنامه جستارهای اقتصادی ایران با رویکرد اقتصاد اسلامی، 18(35)، 169-191.
5. سلامی، حبیب اله و مافی، حسن. (1397). پیشبینی قیمت صادراتی پسته ایران مبتنی بر چرخههای تجاری: کاربست الگوی سریزمانی ساختاری. تحقیقات اقتصاد و توسعه کشاورزی ایران، 49(4)، 571-559.
6. شیرانی فخر، زهره، خوش اخلاق، رحمان، و شریفی، علی مراد. (1393). تخمین تابع تقاضای گاز طبیعی بخش صنعت ایران با استفاده از مدل سری زمانی ساختاری. مطالعات اقتصادی کاربردی ایران (مطالعات اقتصادی کاربردی)، 3(11)، 129-157.
7. عباسیان، عزت ا...، محمودی، وحید و شاکر، ایمان. (1390). تحلیل خطای پیشبینی درآمدهای مالیاتی دولت در اقتصاد ایران. تحقیقات مالی، 13(32)، 109-132.
8. عزتی شورگلی، احمد، زینالزاده شیخ سرمست، مهتاب و عیسوی، هیرو. (1402)، پیشبینی اثر وصل شدن تراکنش دستگاه کارتخوانها به سیستم سازمان امور مالیاتی در افزایش وصولی سازمان امور مالیاتی استان آذربایجان غربی. پژوهشنامه مالیات، ۳۴(۶۰)، ۱۳۴-۱۶۵.
9. نصراللهی، زهرا و شاکر اردکانی، ایمان. (1395). بررسی عملکرد طراحان بودجه در پیشبینی درآمدهای دولت در اقتصاد ایران. پژوهشهای اقتصادی (رشد و توسعه پايدار)، ۱۶(۴)، ۱۲۷-۱۴۷.
10. References
11. Abbasian, E., Mahmoudi, V., & Shaker, I. (2012). Forecast Error Analysis of State Tax Revenues in Iran. Financial Research Journal, 13(32), 109-132. [In Persian]
12. Abdi, M., Hamidialamdari, S., & Pourhassanamiri, M. (2011). An Evaluation of Forecasting Methods and Optimal Combination Models to Predict Tax Revenues. Journal of Tax Research, 19(11), 85-120. [In Persian]
13. Altaf, S., Khan, S. Y., Ali, M. W., & Khan, M. W. (2021). A mathematical forecasting model to estimate tax to GDP ratio. International Journal of Advanced Research in Enginering and Technology, 12(3), 291-297.
14. Andrejovska, A., & Pulikova, V. (2018). Tax revenues in the context of economic determinants. Montenegrin Journal of Economics, 14(1), 133-141. [DOI:10.14254/1800-5845/2018.14-1.10]
15. Arfaee, M., Bahari, A., & Khalilzadeh, M. (2022). A novel prediction model for educational planning of human resources with data mining approach: a national tax administration case study. Education and Information Technologies, 27(2), 2209-2239. [DOI:10.1007/s10639-021-10699-6]
16. Battiston, P., Gamba, S., & Santoro, A. (2024). Machine learning and the optimization of prediction-based policies. Technological Forecasting and Social Change, 199, 123080. [DOI:10.1016/j.techfore.2023.123080]
17. Buettner, T., & Kauder, B. (2010). Revenue forecasting practices: Differences across countries and consequences for forecasting performance. Fiscal Studies, 31(3), 313-340. [DOI:10.1111/j.1475-5890.2010.00117.x]
18. Castro, G. Á., & Camarillo, D. B. R. (2014). Determinants of tax revenue in OECD countries over the period 2001-2011. Contaduría Y Administración, 59(3), 35-59. [DOI:10.1016/S0186-1042(14)71265-3]
19. Chatzis, S. P., Siakoulis, V., Petropoulos, A., Stavroulakis, E., & Vlachogiannakis, N. (2018). Forecasting stock market crisis events using deep and statistical machine learning techniques. Expert Systems with Applications, 112, 353-371. [DOI:10.1016/j.eswa.2018.06.032]
20. Dabbagh, R., Golmoradi, H., & Bagri, A. (2019). Financial performance comparison of Islamic and conventional banking in selected countries using the CAMEL model. Quarterly Studies in Banking Management and Islamic Banking, 4 (Autumn), 85-114. [In Persian]
21. Dabbagh, R., Farzan, M. and Mohseni, B. (2025). Identifying the Challenges of Industry 4.0 Technologies for Sustainable Operations in SMEs. Research in Production and Operations Management, 16(1), 1-24. [In Persian]
22. Ezzati Shourghouli, A., Zeinalzadeh, M., & Isavi, H. (2024). Predicting the effect of connecting the transaction of card readers to the system of the tax affairs organization in increasing the collection of the tax affairs organization of West. J Tax Res, 31 (60), 133-165. [In Persian] [DOI:10.61186/taxjournal.34.60.134]
23. Febriminanto, R. D., & Wasesa, M. (2022). Machine Learning Analytics for Predicting Tax Revenue Potential. Indonesian Treasury Review: Jurnal Perbendaharaan, Keuangan Negara dan Kebijakan Publik, 7(3), 193-205. [DOI:10.33105/itrev.v7i3.497]
24. Fikriya, A. A., & Hikmawati, S. (2020). Support Vector Machine Predictive Analysis Implementation: Case Study of Tax Revenue in Government of South Lampung. In Proceeding International Conference on Science and Engineering (Vol. 3, pp. 323-327). [DOI:10.14421/icse.v3.521]
25. Hajej, Z., Rezg, N., & Gharbi, A. (2017). Ecological optimization for forecasting production and maintenance problem based on carbon tax. The International Journal of Advanced Manufacturing Technology, 88, 1595-1606. [DOI:10.1007/s00170-016-8877-9]
26. Holkova, B., Malichova, E., Falat, L., & Pancikova, L. (2023). Determinants of tax ethics in society: Statistical and logistic regression approach. Mathematics, 11(10), 2341. [DOI:10.3390/math11102341]
27. Huang, W., Lai, K. K., Nakamori, Y., Wang, S., & Yu, L. (2007). Neural networks in finance and economics forecasting. International Journal of Information Technology & Decision Making, 6(01), 113-140. [DOI:10.1142/S021962200700237X]
28. Kalaš, B., Todorović, J. Đ., & Đorđević, M. (2020). Panel estimating effects of macroeconomic determinants on tax revenue level in European :union:. Industrija, 48(3). [DOI:10.5937/industrija48-27820]
29. Karpowicz, A., & Majewska, E. (2018). Corporate Income Tax Revenue Determinants: How Important is the Tax Rate? Economic and Social Development: Book of Proceedings, 361-369.
30. Koirala, T.P. (2012) Government Revenue Forecasting in Nepal; NRB Economic Review, Vol. 24: 47-60. [DOI:10.3126/nrber.v24i2.52727]
31. Kuan, C. M., & White, H. (1994). Artificial neural networks: An econometric perspective. Econometric Reviews, 13(1), 1-91. [DOI:10.1080/07474939408800273]
32. Lahiri, K., & Yang, C. (2022). Boosting tax revenues with mixed-frequency data in the aftermath of COVID-19, The case of New York. International Journal of Forecasting, 38(2), 545-566. [DOI:10.1016/j.ijforecast.2021.10.005]
33. Lin, C. H., Lin, I. C., Wu, C. H., Yang, Y. C., & Roan, J. (2012). The application of decision tree and artificial neural network to income tax audit: the examples of profit-seeking enterprise income tax and individual income tax in Taiwan. Journal of the Chinese Institute of Engineers, 35(4), 401-411. [DOI:10.1080/02533839.2012.655901]
34. Li-Xia, L., Yi-Qi, Z., & Liu, X. Y. (2011). Tax forecasting theory and model based on SVM optimized by PSO. Expert Systems with Applications, 38(1), 116-120. [DOI:10.1016/j.eswa.2010.06.022]
35. Ma, L., Shi, M., Li, Y., & Liu, Y. (2019). Prediction of Total Tax Value Based on Grey BP Correction Model. Journal of Economics, Business and Management, 7(2), 45-49. [DOI:10.18178/joebm.2019.7.2.579]
36. Malikov, T. (2021). Methodological approaches to assessing and forecasting the tax potential of the region. Turkish Journal of Computer and Mathematics Education (TURCOMAT), 12(11), 7056-7060.
37. Maulia, E., & Sofyan, H. (2018). Tax revenue and inflation rate predictions in Banda Aceh using Vector Error Correction Model (VECM). In IOP Conference Series: Materials Science and Engineering (Vol. 352, No. 1, p. 012056). IOP Publishing. [DOI:10.1088/1757-899X/352/1/012056]
38. Milenković, N., Kalaš, B., Mirović, V., & Andrašić, J. (2020). The impact of macroeconomic determinants and tax form on inflation in selected Balkan countries. Serbian Journal of Management, 15(1), 7-18. [DOI:10.5937/sjm15-16685]
39. Molapo, M. A., Olaomi, J. O., & Ama, N. O. (2019). Bayesian vector auto-regression method as an alternative technique for forecasting South African tax revenue. Southern African Business Review, 23(1). [DOI:10.25159/1998-8125/4416]
40. Murorunkwere, B. F., Haughton, D., Nzabanita, J., Kipkogei, F., & Kabano, I. (2023). Predicting tax fraud using supervised machine learning approach. African Journal of Science, Technology, Innovation and Development, 15(6), 731-742. [DOI:10.1080/20421338.2023.2187930]
41. Nasrollahi, Z., & Shaker Ardakani, I. (2016). Investigating the Performance of Budget Designers in Forecasting Government Revenues in the Iranian Economy. QJER, 16(4), 127-147. [In Persian]
42. Neog, Y., & Gaur, A. K. (2020). Macro-economic determinants of tax revenue in India: an application of dynamic simultaneous equation model. International Journal of Economic Policy in Emerging Economies, 13(1), 13-35. [DOI:10.1504/IJEPEE.2020.106679]
43. Nosratabadi, S., Mosavi, A., Duan, P., Ghamisi, P., Filip, F., Band, S. S., & Gandomi, A. H. (2020). Data science in economics: comprehensive review of advanced machine learning and deep learning methods. Mathematics, 8(10), 1799. [DOI:10.3390/math8101799]
44. Pradhan, R. P., Arvin, M. B., Nair, M., Bennett, S. E., & Bahmani, S. (2024). Some determinants and mechanics of economic growth in middle-income countries: The role of ICT infrastructure development, taxation and other macroeconomic variables. The Singapore Economic Review, 69(01), 297-333. [DOI:10.1142/S0217590820500563]
45. Ruhi, M., Abbasian, E., Momeni, M., & Amouzad, M., (2017). Prediction of corporate income tax collection in Iran, using Markov chain and discrete spectrum analysis. J Tax Res, 26(38), 107-129. [In Persian].
46. Saadat Mehr, M. (2021). Investigation of the impact of macroeconomic variables on government tax revenues in Iran. Journal of Iran's Economic Essays (JIEE), 18(35), 169-191. [In Persian].
47. Salami, H., & Mafi, H. (2018). Predicting Export prices of the Iranian Pistachio Based on Commercial Cycles: Application of Structural Time Series Model. Iranian Journal of Agricultural Economics and Development Research, 49(4), 559-571. [In Persian].
48. Sevim, C., Oztekin, A., Bali, O., Gumus, S., & Guresen, E. (2014). Developing an early warning system to predict currency crises. European Journal of Operational Research, 237(3), 1095-1104. [DOI:10.1016/j.ejor.2014.02.047]
49. Shiranifakhr, Z., Khoshakhlagh, R., & Sharifi, A. M. (2014). Estimating Demand Function for Natural Gas in the Industrial Sector of Iran using Structural Time Series Model (STSM). Journal of Applied Economics Studies in Iran, 3(11), 129-157. [In Persian].
50. Thayyib, P. V., Thorakkattle, M. N., Usmani, F., Yahya, A. T., & Farhan, N. H. (2023). Forecasting Indian Goods and Services Tax revenue using TBATS, ETS, Neural Networks, and hybrid time series models. Cogent Economics & Finance, 11(2), 2285649. [DOI:10.1080/23322039.2023.2285649]
51. Xu, C., & Kong, Y. (2024). Random forest model in tax risk identification of real estate enterprise income tax. Plos One, 19(3), e0300928. [DOI:10.1371/journal.pone.0300928]
52. Zareie, P., Jalaee, S. A., & Sadeghi, Z. (2019) Simulation and prediction of the green tax effect on energy consumption and intensity in Iran using a genetic algorithm. J Tax Res, 27(42), 103-125. [In Persian]. [DOI:10.29252/taxjournal.27.42.103]
53. Zhang, D., Zheng, S., & Fu, W. (2023). Research on the prediction model of Chinese tax revenue based on GM (1, 1) and LSSVM. Information Technology and Control, 52(4), 811-818. [DOI:10.5755/j01.itc.52.4.32693]
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
