Неформальная экономика в Центральной Африке и Средиземноморье: анализ на основе машинного обучения

Неформальная экономика остается устойчивым и многомерным явлением, особенно в развивающихся регионах, где ее динамика определяется сочетанием институциональных ограничений и внешних шоков.

Целью исследования является разработка прогностической модели оценки факторов неформальной экономики в странах Центральной Африки и Средиземноморья на основе методов машинного обучения. Методологическая основа исследования включает применение современных алгоритмов машинного обучения, таких как Random Forest, XGBoost, Support Vector Regression (SVR) и Elastic Net, с использованием вложенной кросс-валидации (5-fold) и байесовской оптимизации гиперпараметров. Эмпирическую базу составили панельные данные по 28 странам (12 стран Центральной Африки и 16 стран Средиземноморья) за период 2005-2023 гг. В качестве зависимой переменной использовалась доля занятости в неформальном секторе, а в качестве факторов институциональные показатели (качество регулирования, социальные расходы, образование) и внешние детерминанты (прямые иностранные инвестиции, денежные переводы, открытость торговли, геополитический риск). Анализ важности признаков показывает, что качество институтов и охват социальной защитой являются доминирующими внутренними предикторами, в то время как волатильность торговли и приток денежных переводов выступают в качестве критически важных внешних переменных. Random Forest (R² = 0,983; MAPE = 2,57%) и SVR (R² = 0,982; MAPE = 2,17%) также подтвердили высокую точность прогнозирования. Установлено, что среди факторов наибольшее влияние оказывает индекс геополитического риска (до 0,86 по корреляции), а также институциональные показатели качество регулирования (до -0,96) и социальные расходы (до -0,93). Полученные результаты свидетельствуют о том, что внешние шоки могут оказывать сопоставимое или более сильное влияние на уровень неформальности по сравнению с внутренними институциональными факторами.

1. Acosta, P. A., Lartey, E. K. K., & Mandelman, F. S. (2009). Remittances and the Dutch disease (Working Paper No. 2007-8a). Federal Reserve Bank of Atlanta. Retrieved March 30, 2026 from https://www.econstor.eu/handle/10419/70606

2. Alises, G. F. P., Avignone, T., & Jiménez, M. T. (2025). Predicting migrant children’s social exclusion risk through an innovative digital tool: Application of machine learning methods to Spanish residential centres. Children and Youth Services Review, 175, 108345.f https://doi.org/10.1016/j.childyouth.2025.108345

3. Apaydın, M., Yumuş, M., Değirmenci, A., & Karal, Ö. (2022). Evaluation of air temperature with machine learning regression methods using Seoul City meteorological data. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 28(5), 737-747.f https://doi.org/10.5505/pajes.2022.66915

4. Arık, D. T., Karal, Ö., & Şahin, A. B. (2020). A Comparative Study of Artificial Neural Networks and Naïve Bayes Techniques for the Classification of Radar Targets. Bitlis Eren Üniversitesi Fen Bilimleri Dergisi, 9(4), 1779-1788. fhttps://doi.org/10.17798/bitlisfen.676973

5. Bedford, K. (2024). Gender and development in the World Bank: An evaluation of the business case for equality. In A. Vetterlein & T. Schmidtke (Eds.), The Elgar companion to the World Bank (pp. 227–238). Edward Elgar Publishing. https://doi.org/10.4337/9781802204780.00032

6. Ben Dalla, L. O. F., Medeni, T. D., Medeni, I. T., & Ulubay, M. (2025). Enhancing Healthcare Efficiency at Almasara Hospital: Distributed Data Analysis and Patient Risk Management. Economy: Strategy and Practice, 19(4), 54–72. https://doi.org/10.51176/1997-9967-2024-4-54-72

7. Ben Dalla, L., Medeni, T. M., Agila, A. A., & Medeni, İ. M. (2024a). Architectural Synergy: Investigating the Role of Artificial Neural Networks in Enabling Deep Learning. The International Journal of Engineering & Information Technology, 12(1), 96-103.f https://doi.org/10.36602/ijeit.v12i1.483

8. Ben Dalla, L., Medeni, T., Zbeida, S., & Medeni, İ. (2024b). Unveiling the evolutionary journey based on tracing the historical relationship between artificial neural networks and deep learning. The International Journal of Engineering & Information Technology (IJEIT), 12(1), 104–110. https://doi.org/10.36602/ijeit.v12i1.484

9. Blumenstock, J. (2019). Fighting Poverty with Data. Science.Retrieved March 30, 2026 from https://www.science.org/doi/10.1126/science.aah5217

10. Borisov, A. N., Borodin, A. I., Gubarev, R. V., Dzuyba, E. I., & Kulikova, O. M. (2024). Assessing the Level of Employment in the Informal Sector of the Economy of Russian Regions Using Modern Machine Learning Methods. Review of Business and Economics Studies, 12(4), 42-57.f https://doi.org/10.26794/2308-944X-2024-12-4-42-57

11. Caldara, D., & Iacoviello, M. (2022). Measuring geopolitical risk. American Economic Review, 112(4), 1194–1225. https://doi.org/10.1257/aer.20191823

12. Chen, M. A. (2012). The informal economy: Definitions, theories and policies (WIEGO Working Paper No. 1). Women in Informal Employment: Globalizing and Organizing (WIEGO). https://www.wiego.org/wp-content/uploads/2019/09/Chen_WIEGO_WP1.pdf

13. Dabla-Norris, E, J Brumby, A Kyobe, Z Mills and C Papageorgiou (2012). Investing in public investment: An index of public investment efficiency. Journal of Economic Growth, 17, 235–266. https://doi.org/10.1007/s10887-012-9078-5

14. Dalla, L. O. B., Karal, Ö., & Degirmenciyi, A. (2025). Leveraging LSTM for Adaptive Intrusion Detection in IoT Networks: A Case Study on the RT-IoT2022 Dataset implemented On CPU Computer Device Machine.f 5th International Conference on Engineering, Natural and Social Sciences, April 15-16, 2025: Konya, Turkey, 2025.

15. De Soto, H. (1989). The other path: The invisible revolution in the third World. New York: Harper & Row. https://pdfs.semanticscholar.org/ed48/ad31ae1779a3dce330c1afdfe152e1f7fc55.pdf

16. Degirmenci, A., & Karal, O. (2021). Robust incremental outlier detection approach based on a new metric in data streams. IEEE Access, 9, 160347–160360. https://doi.org/10.1109/ACCESS.2021.3131402

17. Degirmenci, A., & Karal, O. (2022a). iMCOD: Incremental multi-class outlier detection model in data streams. Knowledge-Based Systems, 258, 109950.f https://doi.org/10.1016/j.knosys.2022.109950

18. Degirmenci, A., & Karal, O. (2022b). US accent recognition using machine learning methods. In 2022 Innovations in Intelligent Systems and Applications Conference (ASYU) (pp. 1-6). IEEE. https://doi.org/10.1109/ASYU56188.2022.9925265

19. Dulkadir, S., Tecimer, H.U., Parlaktürk, F., Altındal Ş. & Karal Ö. (2020). The effect of radiation on the forward and reverse bias current–voltage (I–V) characteristics of Au/(Bi4Ti3O12/SiO2)/n-Si (MFIS) structures. Journal of Materials Science: Materials in Electronics, 31, 12514–12521. https://doi.org/10.1007/s10854-020-03801-0

20. Duplock, R., Casali, G. L., & McLennan, C. L. (2025). Predicting Entrepreneurial Intentions and Behaviour: A Machine Learning Approach Using Latent Constructs from Australian GEM Data. Sage Open, 15(4), 21582440251404789. fhttps://doi.org/10.1177/21582440251404789

21. Erkkilä, T., & Piironen, O. (2014). (De) politicizing good governance: The World Bank Institute, the OECD and the politics of governance indicators. Innovation: The European Journal of Social Science Research, 27(4), 344-360. https://doi.org/10.1080/13511610.2013.850020

22. Faraj, L. O. (2017). Observations on evolution of lean software development [Yalın yazılım geliştirme süreci üzerine gözlemler] [Master's thesis, Atilim University]. Council of Higher Education National Thesis Center. Retrieved March 30, 2026 from https://tez.yok.gov.tr/UlusalTezMerkezi/tezDetay.jsp?id=R_EJxYiWWNffOuWM4F4eXQ&no=fiwArXgOvJPKmFC-nX3H-w

23. Farjallah, N. (2025). Machine Learning-Based Modeling of Economic Growth and Governance Quality: The MENA region. Journal of Cultural Analysis and Social Change, 1003-1011.f https://doi.org/10.64753/jcasc.v10i3.2536

24. Felix, J., Alexandre, M., & Lima, G. T. (2025). Applying machine learning algorithms to predict the size of the informal economy. Computational Economics, 65(3), 1169-1189.f https://doi.org/10.1007/s10614-024-10593-6

25. Friedman, E., Johnson, S., Kaufmann, D. and Zoido-Lobaton, P. (2000). Dodging the grabbing hand: the determinants of unofficial activity in 69 countries. Journal of Public Economics, 76(3), pp. 459-93. https://doi.org/10.1016/S0047-2727(99)00093-6

26. Gammeltoft, P. (2002). Remittances and other financial flows to developing countries. International migration, 40(5), 181-211.f https://doi.org/10.1111/1468-2435.00216

27. Heyneman, S. P. (1999). The sad story of UNESCO’s education statistics. International Journal of Educational Development, 19(1), 65–74. https://comparative-education.com/wp-content/uploads/2018/03/the-sad-story-of-unescos-education-statistics.pdf

28. Yalman, Y., Uyanık, T., Atli, I., Tan, A., Bayındır, K.Ç., Karal, Ö., Golestan, S., & Guerrero, J.M. (2022). Prediction of Voltage Sag Relative Location with Data-Driven Algorithms in Distribution Grid. Energies, 15(18), 6641. https://doi.org/10.3390/en15186641

29. Iftikhar, M. N., Justice, J. B., & Audretsch, D. B. (Eds.). (2020). Urban studies and entrepreneurship. Springer. https://doi.org/10.1007/978-3-030-15164-5

30. ILO. (2021). World Employment and Social Outlook: Trends 2021. Retrieved March 30, 2026 from https://www.ilo.org/wcmsp5/groups/public/@dgreports/@dcomm/@publ/documents/publication/wcms_795453.pdf

31. Karal, Ö. (2018). Destek vektör regresyon ile EKG verilerinin sıkıştırılması [Compression of ECG data by support vector regression method]. Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi, 33(2), 743– 756. https://doi.org/10.17341/gazimmfd.416527

32. Karal, Ö. (2020). Performance comparison of different kernel functions in SVM for different k value in k-fold cross-validation. In 2020 Innovations in Intelligent Systems and Applications Conference (ASYU) (pp. 1–5). IEEE. https://doi.org/10.1109/ASYU50717.2020.9259880

33. Kotzinos, A., Canellidis, V., & Psychoyios, D. (2023). Informal sector, ICT dynamics, and the sovereign cost of debt: A machine learning approach. Computation, 11(5), 90. https://doi.org/10.3390/computation11050090

34. Loayza, N. V. (1996). The economics of the informal sector: a simple model and some empirical evidence from Latin America. In Carnegie-Rochester conference series on public policy, 45, 129-162. North-Holland.fhttps://doi.org/10.1016/S0167-2231(96)00021-8

35. Muttaqi, M., Degirmenci, A., & Karal, O. (2022). US accent recognition using machine learning methods. In 2022 Innovations in Intelligent Systems and Applications Conference (ASYU), 1-6. IEEE. https://doi.org/10.1109/ASYU56188.2022.9925265

36. Osunnaiye, A. V., & Kucukaltan, B. (2025). Examining the influence of sustainable development indicators on economic development: a machine learning approach with evidence from Africa. International Journal of Productivity and Performance Management, 74(9), 3131-3152.f https://doi.org/10.1108/IJPPM-12-2024-0839

37. Salman, M., & Wang, G. (2025). Analysis of natural resource efficiency convergence in global North and South: The role of artificial intelligence and geopolitical risks in club formation. Environment, Development and Sustainability, 28, 7937–7969.f https://doi.org/10.1007/s10668-025-06863-4

38. Seyid A., E. A., Bounabi, M., Azmi, R., Diop, E. B., Hlal, M., Almouctar, M. A. S., Chenal, J., & Adraoui, M. (2025). Forecasting urban water demand in Ben Guerir Morocco using statistical and machine learning methods. Discover Sustainability, 6(1), 1-200. https://doi.org/10.1007/s43621-025-02086-9

39. Soliman, A., Shlibak, A., & Zencirci, N. (2026). Wheat Fungal Diseases: A Review. Wadi Alshatti University Journal of Pure and Applied Sciences, 4(1), 191-198. https://doi.org/10.63318/waujpasv4i1_20

40. UNCTAD. (2020). Impact of the coronavirus outbreak on global FDI (Investment Trends Monitor, Special Issue). Retrieved March 30, 2026 from https://unctad.org/press-material/impact-coronavirus-outbreak-global-fdi