SavedAdd your journal

Journals / InTraders Uluslararası Ticaret Akademik Dergisi / 2026 / Volume: 9 Issue: 1

Forecasting Turkish Exports: A Comparative Study of Statistical, Machine Learning, and Deep Learning Methods

Ümit Yılmaz(BALIKESİR ÜNİVERSİTESİ, BİGADİÇ MESLEK YÜKSEKOKULU, YÖNETİM VE ORGANİZASYON BÖLÜMÜ, LOJİSTİK PR.)0000-0003-4268-8598
Fehime Günbegi(BALIKESIR UNIVERSITY, BİGADİÇ VOCATIONAL SCHOOL, DEPARTMENT OF FINANCE, BANKING AND INSURANCE, BANKING AND INSURANCE PR.)0000-0003-1410-4452
Pages
—
Publication Date
2026-07-30
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Results indicate that the SARIMAX+CatBoost hybrid model achieves the highest predictive accuracy (R² = 0.857, MAPE = 5.67%), marginally outperforming standalone SARIMAX (R² = 0.853, MAPE = 5.72%).

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Accurate export forecasting is a critical input for macroeconomic policy formulation, trade strategy design, and fiscal planning.
📊
Key Findings & Data:This study presents a systematic comparison of eight forecasting models, namely SARIMAX, Prophet, Hybrid-S (SARIMAX+CatBoost), Hybrid-P (Prophet+CatBoost), XGBoost, CatBoost, Gated Recurrent Unit (GRU), and Neural Basis Expansion Analysis for Time Series (N-BEATS), applied to monthly Turkish export data spanning January 2002 to December 2025, comprising 288 observations.
💡
Conclusion & Impact:Results indicate that the SARIMAX+CatBoost hybrid model achieves the highest predictive accuracy (R² = 0.857, MAPE = 5.67%), marginally outperforming standalone SARIMAX (R² = 0.853, MAPE = 5.72%).
🔬Design:Modeling & Simulation
📍Scope:2002 Verileri
⚙️Variables:Export forecasting · hybrid model · walk-forward validation

Abstract

Accurate export forecasting is a critical input for macroeconomic policy formulation, trade strategy design, and fiscal planning. This study presents a systematic comparison of eight forecasting models, namely SARIMAX, Prophet, Hybrid-S (SARIMAX+CatBoost), Hybrid-P (Prophet+CatBoost), XGBoost, CatBoost, Gated Recurrent Unit (GRU), and Neural Basis Expansion Analysis for Time Series (N-BEATS), applied to monthly Turkish export data spanning January 2002 to December 2025, comprising 288 observations. Exogenous variables include the Brent crude oil price, the real effective exchange rate, and the industrial production index. Walk-forward expanding-window validation is employed to ensure that each one-step-ahead forecast is generated exclusively from historically available information, thereby eliminating look-ahead bias; hyperparameters are optimized via Optuna with 50 trials; and statistical significance is assessed using Diebold-Mariano tests. Results indicate that the SARIMAX+CatBoost hybrid model achieves the highest predictive accuracy (R² = 0.857, MAPE = 5.67%), marginally outperforming standalone SARIMAX (R² = 0.853, MAPE = 5.72%). Deep learning methods (GRU, N-BEATS) perform substantially below all other models, achieving R² values of 0.44 and 0.39 respectively, consistent with the broader literature finding that neural networks underperform on low-frequency macroeconomic series with structural breaks and limited sample sizes. This study provides the first unified eight-model benchmark for Turkish export forecasting under walk-forward validation, introduces two novel hybrid architectures validated across multiple structural regimes including the COVID-19 shock, and empirically confirms the data-regime hypothesis: statistical and hybrid methods dominate in low-frequency, small-sample, structurally unstable macroeconomic environments.

🔖

Cite This Article

Yılmaz, Ü., Günbegi, F. (2026). Forecasting Turkish Exports: A Comparative Study of Statistical, Machine Learning, and Deep Learning Methods. InTraders Uluslararası Ticaret Akademik Dergisi, Volume: 9 Issue: 1. https://doi.org/10.55065/intraders.1905139

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (42)

  1. Akiba, T., Sano, S., Yanase, T., Ohta, T., & Koyama, M. (2019). Optuna: A Next-generation Hyperparameter Optimization Framework Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Anchorage, AK, USA. https://doi.org/10.1145/3292500.3330701
  2. Alakbarov, N., Gündüz, M., & Özkaya, M. H. (2017). Türkiye’nin İhracat Talebi Fonksiyonunun Farklı Ülke Gruplarına Göre Eşbütünleşme Analizi. Dokuz Eylül Üniversitesi İktisadi İdari Bilimler Fakültesi Dergisi, 32(2), 369–395. https://doi.org/10.24988/deuiibf.2017322662
  3. Altintaş, H., Cetin, R., & Öz, B. (2011). The Impact of Exchange Rate Volatility on Turkish Exports: 1993-2009. South East European Journal of Economics and Business, 6(2), 71–81. https://doi.org/10.2478/v10033-011-0017-8
  4. Avşar, İ. İ., & Ecemiş, O. (2023). Türkiye’nin İthalat ve İhracatının Veri Madenciliği Yöntemleriyle Tahmini. Osmaniye Korkut Ata Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 6(3), 1890–1907. https://doi.org/10.47495/okufbed.1220874
  5. Bates, J. M., & Granger, C. W. J. (1969). The Combination of Forecasts. OR, 20(4), 451–468. https://doi.org/10.2307/3008764
  6. Box, G. E. P., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2016). Time series analysis: Forecasting and control. John Wiley & Sons. https://www.wiley.com/en-us/Time+Series+Analysis%3A+Forecasting+and+Control%2C+5th+Edition-p-9781118675021
  7. Chen, T., & Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, California, USA. https://doi.org/10.1145/2939672.2939785
  8. Cho, K., van Merriënboer, B., Bahdanau, D., & Bengio, Y. (2014). On the Properties of Neural Machine Translation: Encoder–Decoder Approaches Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation, Doha, Qatar. https://doi.org/10.3115/v1/W14-4012
  9. Çakmaklı, B. M. (2025). Inflation, Exchange Rate and Foreign Trade: Long And Short Run Relationships In The Turkish Economy. InTraders International Trade Academic Journal, 8(1), 1–19. https://doi.org/10.55065/intraders.1660803
  10. Dave, E., Leonardo, A., Jeanice, M., & Hanafiah, N. (2021). Forecasting Indonesia Exports using a Hybrid Model ARIMA-LSTM. Procedia Computer Science, 179, 480–487. https://doi.org/10.1016/j.procs.2021.01.031
  11. Diebold, F. X., & Mariano, R. S. (1995). Comparing Predictive Accuracy. Journal of Business & Economic Statistics, 13(3), 253–263. https://doi.org/10.2307/1392185
  12. FRED. (2026). Crude Oil Prices: Brent - Europe (DCOILBRENTEU). Federal Reserve Bank of St. Louis. Retrieved 27.01.2026 from https://fred.stlouisfed.org/series/DCOILBRENTEU
  13. Ghauri, S. P., Ahmed, R. R., Streimikiene, D., & Streimikis, J. (2020). Forecasting exports and imports by using autoregressive (AR) with seasonal dummies and Box-Jenkins approaches: A case of Pakistan. Engineering economics, 31(3), 291–301. https://doi.org/10.5755/j01.ee.31.3.25323
  14. Goldstein, M., & Khan, M. (1985). Income and price effects in foreign trade. In R. W. Jones & P. B. Kenen (Eds.), Handbook of International Economics (Vol. 2, pp. 1041–1105). Elsevier. https://doi.org/10.1016/S1573-4404(85)02011-1
  15. Göze Kaya, D., & Durgun Kaygısız, A. (2023). Türkiye’de Bütçe Dengesi ve Enflasyon İlişkisi: Son 20 Yılın Ampirik Analizi [The Relationship Between Budget Balance And Inflation in Turkey: An Empirical Analysis of The Last 20 Years]. InTraders International Trade Academic Journal, 6(2), 201–222. https://doi.org/10.55065/intraders.1395563
  16. Gu, K., Cao, L., Cao, J., Li, M., Chen, H., Wang, Z., Liu, S., & Zhang, K. (2026). A hybrid SVMD-RIME-TCN-BiGRU model for wind power prediction. Energy Informatics, 9(1), 8. https://doi.org/10.1186/s42162-026-00630-z
  17. Gulzar, Y., Oral, C., Kayakus, M., Erdogan, D., Unal, Z., Eksili, N., & Caylak, P. C. (2024). Predicting High Technology Exports of Countries for Sustainable Economic Growth by Using Machine Learning Techniques: The Case of Turkey. Sustainability, 16(13), 5601. https://doi.org/10.3390/su16135601
  18. Harb, N. (2007). Trade between Euro zone and Arab countries: a panel study. Applied Economics, 39(16), 2099–2107. https://doi.org/10.1080/00036840600722307
  19. Hewamalage, H., Bergmeir, C., & Bandara, K. (2021). Recurrent Neural Networks for Time Series Forecasting: Current status and future directions. International Journal of Forecasting, 37(1), 388–427. https://doi.org/10.1016/j.ijforecast.2020.06.008
  20. Iossifov, P., & Fei, X. (2019). Real Effective Exchange Rate and Trade Balance Adjustment. IMF Working Papers, 19(131), 1. https://doi.org/10.5089/9781498312738.001
  21. Kavacık, M. (2023). What did Türkiye do in terms of export competitiveness after 2000? InTraders International Trade Academic Journal, 6(2), 119–149. https://doi.org/10.55065/intraders.1390704
  22. Koç Yurtkur, A., & Tay Bayramoğlu, A. (2012). Gelişmekte Olan Piyasa Ekonomilerinde İhracatın Dinamikleri. Yönetim ve Ekonomi Dergisi, 19(1), 19–33. https://izlik.org/JA34PY52LN
  23. Koççat, H. (2008). Exchange rates, exports and economic growth in Turkey: evidence from Johansen cointegration tests. Journal of Economic & Management Perspectives, 2(1), 5. https://www.proquest.com/docview/228313408/fulltextPDF/DFAEEEDCF9BA4F82PQ/
  24. Kumar, S. (2009). An empirical evaluation of export demand in China. Journal of Chinese Economic and Foreign Trade Studies, 2(2), 100–109. https://doi.org/10.1108/17544400910966077
  25. Lehmann, R. (2021). Forecasting exports across Europe: What are the superior survey indicators? Empirical Economics, 60(5), 2429–2453. https://doi.org/10.1007/s00181-020-01838-y
  26. Makridakis, S., Spiliotis, E., & Assimakopoulos, V. (2020). The M4 Competition: 100,000 time series and 61 forecasting methods. International Journal of Forecasting, 36(1), 54–74. https://doi.org/10.1016/j.ijforecast.2019.04.014
  27. Makridakis, S., Spiliotis, E., & Assimakopoulos, V. (2022). M5 accuracy competition: Results, findings, and conclusions. International Journal of Forecasting, 38(4), 1346–1364. https://doi.org/10.1016/j.ijforecast.2021.11.013
  28. Oreshkin, B. N., Carpov, D., Chapados, N., & Bengio, Y. (2019). N-BEATS: Neural basis expansion analysis for interpretable time series forecasting. ArXiv, abs/1905.10437. https://doi.org/10.48550/arXiv.1905.10437
  29. Petropoulos, F., Apiletti, D., Assimakopoulos, V., Babai, M. Z., Barrow, D. K., Ben Taieb, S., Bergmeir, C., Bessa, R. J., Bijak, J., Boylan, J. E., Browell, J., Carnevale, C., Castle, J. L., Cirillo, P., Clements, M. P., Cordeiro, C., Cyrino Oliveira, F. L., De Baets, S., Dokumentov, A.,…Ziel, F. (2022). Forecasting: theory and practice. International Journal of Forecasting, 38(3), 705–871. https://doi.org/10.1016/j.ijforecast.2021.11.001
  30. Piñeres, S. A. G. d., & Cantavella-Jordá, M. (2010). Short-run effects of devaluation: a disaggregated analysis of Latin American exports. Applied Economics, 42(2), 133–142. https://doi.org/10.1080/00036840701579259
  31. Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V., & Gulin, A. (2018). CatBoost: unbiased boosting with categorical features 32nd Conference on Neural Information Processing Systems (NeurIPS 2018), Montréal, Canada. https://papers.nips.cc/paper_files/paper/2018/file/14491b756b3a51daac41c24863285549-Paper.pdf
  32. Rahmaddi, R., & Ichihashi, M. (2012). How Do Foreign and Domestic Demand Affect Exports Performance? An Econometric Investigation of Indonesia's Exports. Modern Economy, 3(1), 32–42. https://doi.org/10.4236/me.2012.31005
  33. Senhadji, A. S., & Montenegro, C. E. (1999). Time Series Analysis of Export Demand Equations: A Cross-Country Analysis. IMF staff papers, 46(3), 259–273. https://doi.org/10.2307/3867643
  34. Shangguan, Z. (2025). An Explainable Machine-Learning Framework Based on XGBoost–SHAP and Big Data for Revealing the Socioeconomic Drivers of Population Urbanization in China. Systems, 13(8), 679. https://doi.org/10.3390/systems13080679
  35. Taylor, S. J., & Letham, B. (2018). Forecasting at Scale. The American Statistician, 72(1), 37–45. https://doi.org/10.1080/00031305.2017.1380080
  36. TCMB. (2026, 2026/01/27). Reel Kesim Güven Endeksi (RKGE) – BIE_RKUFEY Serisi. Türkiye Cumhuriyet Merkez Bankası (TCMB). Retrieved 27.01.2026 from https://evds3.tcmb.gov.tr/tumSeriler/2504/bie_rkufey
  37. TİM. (2026). İhracat Rakamları. Türkiye İhracatçılar Meclisi (TİM). Retrieved 27.01.2026 from https://tim.org.tr/tr/ihracat-rakamlari
  38. TÜİK. (2026, 2026/01/27). Basın Bülteni No: 58192. Türkiye İstatistik Kurumu (TÜİK). https://veriportali.tuik.gov.tr/tr/press/58192
  39. Unal, Seyma N., & Karamelikli, H. (2026). A Comparative Forecasting Framework for Turkey–Germany Trade: Evidence From Time Series and Artificial Neural Networks Models. Journal of Forecasting, 45(3), 1177–1187. https://doi.org/10.1002/for.70084
  40. Vergil, H. (2010). Türkiye ticaretinde ulusal pazar etkisi. İktisat İşletme ve Finans 25(286), 35–59. https://doi.org/10.3848/iif.2010.286.2416
  41. Wong, H.-L., Tu, Y.-H., & Wang, C.-C. (2010). Application of fuzzy time series models for forecasting the amount of Taiwan export. Expert Systems with Applications, 37(2), 1465–1470. https://doi.org/10.1016/j.eswa.2009.06.106
  42. Yuan, Y. (2017). Forecasting method for import and export trade on the basis of GMDH network model. Journal of Discrete Mathematical Sciences and Cryptography, 20(4), 755–766. https://doi.org/10.1080/09720529.2017.1358859