Research Article

Comparing machine learning and classical methods in forecasting Türkiye’s macroeconomic performance

Volume: 23 Number: 2026 August 31, 2026
TR EN

Comparing machine learning and classical methods in forecasting Türkiye’s macroeconomic performance

Abstract

This study forecasts Türkiye’s medium-term macroeconomic performance through a composite index based on growth, unemployment, inflation, the budget balance, and the current account balance. Quarterly data for 2006Q1–2026Q1 are used to compare artificial neural networks, ARIMA/SARIMA models, and ordinary least squares regression. Model performance is evaluated through rolling-origin validation over 40 out-of-sample periods using MAE, MAPE, and RMSE. The statistical significance of forecast error differences is examined with the Diebold–Mariano test, while the sensitivity of the neural network is assessed across 180 hyperparameter configurations. The results show that the neural network produces the lowest error in the baseline specification, although its advantage is not statistically significant at the 5 percent level. When seasonal information and lagged component values are included, OLS yields the lowest forecast error. Conditional forecasts from the preferred OLS specification place the index between 91.66 and 94.56 during 2026Q2–2028Q1. The projected path remains broadly stable, with seasonal fluctuations but no pronounced upward or downward trend. Overall, the findings indicate that forecast performance depends on model specification and that claims of machine learning superiority should be evaluated cautiously.

Keywords

Fiscal policy, budget balance, forecast, artificial neural networks, ARIMA, OLS

References

  1. Afonso, A., & Jalles, J. T. (2013). Growth and productivity: The role of government debt. International Review of Economics & Finance, 25, 384–407.
  2. Armstrong, J. S. (Ed.). (2001). Principles of forecasting: A handbook for researchers and practitioners. Springer. Beck, N., Jonathan, N., Alvarez, M., Garrett, G., & Lange, P. (1993). Government partisanship, labor organization, and macroeconomic performance: A corrigendum. American Political Science Review, 87(4), 943–948.
  3. Bishop, C. M. (1995). Neural networks for pattern recognition. Oxford University Press.
  4. Calmfors, L., & Driffill, J. (1988). Bargaining structure, corporatism and macroeconomic performance. Economic Policy, 3(6), 13–61.
  5. Chai, T., & Draxler, R. R. (2014). Root mean square error (RMSE) or mean absolute error (MAE)? Arguments against avoiding RMSE in the literature. Geoscientific Model Development, 7(3), 1247–1250.
  6. Coulombe, G., Leroux, M., Stevanovic, D., & Surprenant, S. (2019). How is machine learning useful for macroeconomic forecasting? Montréal, Canada: CIRANO.
  7. Dağbaşı, B., Barak, D., & Çelik, T. (2019). Türkiye için makroekonomik performans endeksinin analizi (1990–2017): Yapay sinir ağı yaklaşımı. Bingöl Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, 3(1), 93–112.
  8. Diebold, F. X. (2015). Forecasting in economics, business, finance and beyond. Journal of Economic Perspectives, 29(2), 145–168.
  9. Dullien, S. (2017). A new magic square for inclusive and sustainable economic growth: A policy framework for Germany to move beyond GDP. Friedrich-Ebert-Stiftung.
  10. Fischer, S. (1993). The role of macroeconomic factors in growth. Journal of Monetary Economics, 32(3), 485–512.
APA
Kıratoğlu, E. (2026). Comparing machine learning and classical methods in forecasting Türkiye’s macroeconomic performance. OPUS Journal of Society Research, 23(2026), 1-14. https://doi.org/10.26466/opusjsr.1825433
AMA
1.Kıratoğlu E. Comparing machine learning and classical methods in forecasting Türkiye’s macroeconomic performance. OPUS JSR. 2026;23(2026):1-14. doi:10.26466/opusjsr.1825433
Chicago
Kıratoğlu, Emrah. 2026. “Comparing Machine Learning and Classical Methods in Forecasting Türkiye’s Macroeconomic Performance”. OPUS Journal of Society Research 23 (2026): 1-14. https://doi.org/10.26466/opusjsr.1825433.
EndNote
Kıratoğlu E (August 1, 2026) Comparing machine learning and classical methods in forecasting Türkiye’s macroeconomic performance. OPUS Journal of Society Research 23 2026 1–14.
IEEE
[1]E. Kıratoğlu, “Comparing machine learning and classical methods in forecasting Türkiye’s macroeconomic performance”, OPUS JSR, vol. 23, no. 2026, pp. 1–14, Aug. 2026, doi: 10.26466/opusjsr.1825433.
ISNAD
Kıratoğlu, Emrah. “Comparing Machine Learning and Classical Methods in Forecasting Türkiye’s Macroeconomic Performance”. OPUS Journal of Society Research 23/2026 (August 1, 2026): 1-14. https://doi.org/10.26466/opusjsr.1825433.
JAMA
1.Kıratoğlu E. Comparing machine learning and classical methods in forecasting Türkiye’s macroeconomic performance. OPUS JSR. 2026;23:1–14.
MLA
Kıratoğlu, Emrah. “Comparing Machine Learning and Classical Methods in Forecasting Türkiye’s Macroeconomic Performance”. OPUS Journal of Society Research, vol. 23, no. 2026, Aug. 2026, pp. 1-14, doi:10.26466/opusjsr.1825433.
Vancouver
1.Emrah Kıratoğlu. Comparing machine learning and classical methods in forecasting Türkiye’s macroeconomic performance. OPUS JSR. 2026 Aug. 1;23(2026):1-14. doi:10.26466/opusjsr.1825433