Modeling the Probability of the Detection Process of Tax Evasion Taking into Account Quality and Quantity Indicators

Main Article Content

Akif Musayev
Mirvari Gazanfarli


Aims: In contrast to the classical approaches of the standard model of tax evasion based on game theory, our manuscript has considered the detection of tax evasion as one of the main function of tax administration and has proposed a model for assessing the probability of tax evasion taking into consideration qualitative and quantitative indicators.

Study Design: This investigation has been carried out on the basis of research methods such as scientific abstraction and systematic analysis, expert evaluation, logical generalization, statistical analysis.

Place and Duration of Study: Department of Mathematical provision of economic researches, between November 2019 and May 2020.

Methodology: For the evaluating of the probability of tax evasion’s detection firstly, efficiency indicators of tax administration were selected including 3 groups such as internal environmental, external micro, and external macro-environmental factors. These indicators consist of both quantitative as well as qualitative indicators. Quantitative indicators were assessed on the base on statistics information base. Quantitative indicators were assessed on the base of expert skills, knowledge, and experiences in accordance with under investigation countries. The objectiveness of obtained data that characterize qualitative indicators was checked and used both these as well as quantitative indicators for formulating the tax efficiency index. The next step is consists of using these formulations for evaluating the probability of detection of tax evasion under uncertainty. The impact degrees (membership functions) of the parameters that characterize the influence of 3 groups-environmental factors, in the detection of tax evasion were defined, and taking them into account in the fuzzy inference system probability of detection of tax evasion was assessed.

Limitations: Lack or uncertainties of the information base cause difficulties in applying our model.

Results: The probability of detection of tax evasion in the Republic of Azerbaijan was assessed with the proposed model and depends on the results recommendations have been consulted for improving appropriate tax system. As a result of model the probability of detection of tax evasion was defined 29%. The result shows that tax administration mechanism in Azerbaijan Republic need to be improved.

Conclusion: Proposed model drives practical significance as a providing effective activity of tax institutions by defining the level of tax administration, as well as, as an impacting remarkably the revenue of state budget by determining the probability of tax evasion's detection.

Probability of detection of tax evasion, tax administration, efficiency index, qualitative and quantitative indicators, expert evaluation, Mamdani inference system.

Article Details

How to Cite
Musayev, A., & Gazanfarli, M. (2020). Modeling the Probability of the Detection Process of Tax Evasion Taking into Account Quality and Quantity Indicators. Asian Journal of Economics, Business and Accounting, 18(4), 28-37.
Short Research Article


Abrantes PC, Ferraz F. Big data applied to tax evasion detection: A systematic review. 2016 International Conference on Computational Science and Computational Intelligence (CSCI); 2016. DOI:10.1109/csci.2016.

Rahayu Abdul Rahman, Suraya Masrom, Normah Omar. Tax avoidance detection based on machine learning of malaysian government-linked companies. International Journal of Recent Technology and Engineering (IJRTE). 2019;8(2S11). Available:

César PL, Rodríguez MJD, Santos SL. Tax fraud detection through neural networks: An application using a sample of personal income taxpayers. Future Internet. 2019;11:86. Published: 30 March 2019. Available: file:///C:/Users/hp/Downloads/futureinternet-11-00086-v2.pdf.

Daniel R, Boris P, Moreno A, Villamil MP, Figueroa C. Tax fraud detection for under-reporting declarations using an unsupervised machine learning approach. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2018;215–222. Available:

Rahimikiaa E, Shapour M, Rahmani T, Ghazanfaria M. Detecting corporate tax evasion using a hybrid intelligent system: A case study of Iran. International Journal of Accounting Information Systems 25. 2017;1-17. Acailable:

Fisman R, Wei SJ. Tax rates and tax evasion: Evidence from “Missing Imports” in China. Journal of Political Economy. 2004;112(2):471–496.

Allingham MG, Sandmo A. Income tax evasion: A theoretical analysis. Journal of Public Economics. North-Holland Publishing Company. 1972;323-338. Available:

Hindriks and Myles. Lectures on public Finance Part 2 Chap 7. 2014 version. Availabvle:

Escobari D. Imperfect detection of tax evasion in a corrupt tax administration. MPRA Paper No. 2012;39198. Available:

Sandmo A. Income tax evasion, labour supply, and the equity-evasion tradeoff. J. Public Econ. 1981;16(3):265-288.

Pencavel JH. A note on income tax evasion, labor supply, and nonlinear tax schedules. J. Public Econ. 1979;12(1):115-124. Available:

Cowell FA."Taxation and labour supply with risky activities. Economica. 1981;48(192):365-379. Available:

Laurence SW.' The desirability of cheating incentives and randomness in the optimal income tax. J. Polit. Econ. 1976;84(6):1343-1352.

Biswas R, Marchese C, Privileggi F. Tax evasion in a principal-agent model with self-protection. POLIS Working Papers 138, Institute of Public Policy and Public Choice – POLIS; 2009. Available:

Sokolovskyi D. Game-theoretic model of tax evasion: analysis of agents’ interaction and optimization of tax burden. MPRA Paper 86415, University Library of Munich, Germany; 2018. Available:

Mirvari Kh. Gazanfarli. The effectiveness model of tax administration process. Scientific reviews of Azerbaijan State University of Economics. 2020;8(8):92-102. ISSN 2306-8426.