Forecasting Trade with Machine Learning: A Gravity Model Analysis of Türkiye and Balkan Countries

Year 2024, Volume: 27 Issue: 2, 746 - 765, 29.11.2024

Haldun Soydal , Mustafa Ay , Sümeyye Koç

https://doi.org/10.29249/selcuksbmyd.1545687

Abstract

This study analyzes Türkiye's trade volume with Balkan countries using the gravity model and aims to compare the performance of machine learning methods in predicting the trade volume between Türkiye and these countries. For this purpose, data from 2004 to 2023 have been utilized. However, due to the dissolution of Serbia and Montenegro into two independent countries in 2006, and Kosovo's unilateral declaration of independence from Serbia in 2008, the data for these three countries exhibit discontinuous and inconsistent patterns, leading to their exclusion from the analysis. The data used in the study includes numerical variables such as the export and import figures between Türkiye and Balkan countries, the GDP of the countries, and the distance between them. The dummy variables include whether the countries share a border, a common language, their landlocked status, and their memberships in the World Trade Organization (WTO) and the European Customs Union. These data have been analyzed using seven different machine learning models to predict the trade volume. The analysis was conducted using four different train-test data splits and cross-validation with 10 iterations. The performance of the applied machine learning models was compared based on MAPE (Mean Absolute Percentage Error) and R2 values. The findings revealed that the best prediction method varies depending on the country analyzed and the train-test split ratios. These differences in results contribute significantly to a better understanding of trade relations between Türkiye and the Balkan countries, forecasting the future course of these relations, and shaping regional economic policies. Therefore, the study is important in determining strategies for the development of trade with these countries.

Keywords

Gravity Model, Gradient Boosting, Bilateral Trade, Decision Tree, Machine Learning

Makine Öğrenimi ile Ticaretin Öngörülmesi: Türkiye ve Balkan Ülkeleri Üzerine Çekim Modeli Analizi

Abstract

Bu çalışmada, Türkiye'nin Balkan ülkeleri ile olan ticaret hacmi çekim modeli kullanılarak analiz edilmiş ve Türkiye ile bu ülkeler arasındaki ticaret hacmini tahmin etmede makine öğrenimi yöntemlerinin performansları karşılaştırılmak istenmiştir. Bu amaçla çalışmada 2004 yılından başlayarak 2023 yılına kadar olan veriler kullanılmıştır. Ancak Sırbistan ve Karadağ’ın 2006 yılında bağımsız iki ülke hâline gelmesi, Kosova’nın da 2008 yılında Sırbistan’dan tek taraflı bağımsızlığını ilan etmesi nedeniyle verilerinin incelenmesi sonucu görülen devamsız ve sağlıksız yapı sebebi ile bu üç ülke analiz dışında bırakılmıştır. Çalışmada kullanılan veriler sayısal değişkenler olarak Türkiye ile Balkan ülkeleri arasındaki ihracat ve ithalat verileri, ülkelerin milli gelirleri, aralarındaki mesafe; kukla değişkenler olarak ise ülkelerin birbirleriyle olan sınırı, ortak dil, ülkelerin karayla çevrililik durumu, Dünya Ticaret Örgütü (DTÖ) üyelikleri ve Avrupa gümrük birliği üyelikleridir. Bu veriler, ticaret hacmini tahmin etmek için yedi farklı makine öğrenmesi modeli ile analiz edilmiştir. Analiz dört farklı eğitim-test veri seti bölünmesi ve 10 farklı iterasyon ile çapraz doğrulama yoluyla uygulanmıştır. Uygulanan makine öğrenmesi modellerinin başarısı MAPE (Ortalama Mutlak Yüzde Hata) ve R2 değerleri üzerinden kıyaslanmıştır. Bulgular, en iyi tahmin yönteminin analize konu ülkeye ve eğitim-test ayrım oranlarına göre değişim gösterdiğini ortaya koymuştur. Sonuçlar arasındaki bu farklılık, Türkiye ve Balkan ülkeleri arasındaki ticaret ilişkilerinin daha iyi anlaşılması, bu ilişkilerin gelecekteki seyrinin tahmin edilmesi ve bölgesel ekonomik politikaların oluşturulmasına önemli katkılar sağlayacaktır. Bu sebeple çalışma, bu ülkelerle olan ticaretin gelişimine yönelik stratejilerin belirlenmesi açısından önemlidir.

Keywords

Çekim Modeli, Gradyan Artırma, İkili Ticaret, Karar Ağacı, Makine Öğrenmesi.

References

• Abel, J. G., & Gietel-Basten, S. (2020). International remittance flows and the economic and social consequences of COVID-19. Environment and Planning A: Economy and Space, 52(8), 1480–1482. https://doi.org/10.1177/0308518X20931111
• Alesina, A., Favero, C., & Giavazzi, F. (2019). Austerity: When it works and when it doesn’t. Princeton University Press. https://doi.org/10.2307/j.ctvc77f4b
• Alpaydin, E. (2014). Introduction to machine learning (3rd ed.). MIT Press. https://books.google.com.tr/books?id=7f5bBAAAQBAJ
• Anderson, J. E., & van Wincoop, E. (2003). Gravity with gravitas: A solution to the border puzzle. American Economic Review, 93(1), 170–192. https://doi.org/10.1257/000282803321455214
• Baier, S. L., & Bergstrand, J. H. (2007). Do free trade agreements actually increase members’ international trade? Journal of International Economics, 71(1), 72–95. https://doi.org/10.1016/j.jinteco.2006.02.005
• Bangdiwala, S. I. (2018). Regression: Simple linear. International Journal of Injury Control and Safety Promotion, 25(1), 113–115. https://doi.org/10.1080/17457300.2018.1426702
• Bansal, M., Goyal, A., & Choudhary, A. (2022). A comparative analysis of K-Nearest Neighbor, Genetic, Support Vector Machine, Decision Tree, and Long Short-Term Memory algorithms in machine learning. Decision Analytics Journal, 3, 100071. https://doi.org/10.1016/j.dajour.2022.100071
• Bansal, M., Prince, Yadav, R., & Ujjwal, P. K. (2020). Palmistry using machine learning and OpenCV. 2020 Fourth International Conference on Inventive Systems and Control (ICISC), 536–539. https://doi.org/10.1109/ICISC47916.2020.9171158
• Bergstrand, J. H. (1985). The gravity equation in international trade: Some microeconomic foundations and empirical evidence. The Review of Economics and Statistics, 67, 474–481. https://doi.org/10.2307/1925976
• Breiman, L. (1996). Bagging predictors. Machine Learning, 24, 123–140. https://doi.org/10.1007/BF00058655
• Breiman, L. (2001). Random forests. Machine Learning, 45, 5–32. https://doi.org/10.1023/A:1010950718922
• Breiman, L., Friedman, J., Stone, C. J., & Olshen, R. A. (1984). Classification and regression trees. Taylor & Francis. https://books.google.com.tr/books?id=JwQx-WOmSyQC
• 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, 785–794. https://doi.org/10.1145/2939672.2939785
• Circlaeys, S., Kanitkar, C., & Kumazawa, D. (2017). Bilateral trade flow prediction. Unpublished Manuscript, Available at http://cs229.stanford.edu/proj2017/final-reports/5240224.pdf
• Erdoğan, S., & Tek, M. (2019). Türkiye ekonomisinin genel görünümü. In M. Alagöz & G. Akar (Eds.), Sektörel Ekonomik Analiz Türkiye (2003-2018) (pp. 1–34). Gazi Kitabevi.
• Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. The Annals of Statistics, 29(5), 1189–1232. https://doi.org/10.1214/aos/1013203451
• Geurts, P., Ernst, D., & Wehenkel, L. (2006). Extremely randomized trees. Machine Learning, 63, 3–42. https://doi.org/10.1007/s10994-006-6226-1
• Goldfarb, A., & Trefler, D. (2018). Artificial intelligence and international trade. The Economics of Artificial Intelligence: An Agenda, 463–492.
• Hastie, T., Tibshirani, R., & Friedman, J. (2009). The elements of statistical learning: Data mining, inference, and prediction (2nd ed.). Springer.
• Helpman, E., Melitz, M., & Rubinstein, Y. (2008). Estimating trade flows: Trading partners and trading volumes. The Quarterly Journal of Economics, 123(2), 441–487. https://doi.org/10.1162/qjec.2008.123.2.441
• Ho, T. K. (1995). Random decision forests. Proceedings of 3rd International Conference on Document Analysis and Recognition, 1, 278–282. https://doi.org/10.1109/ICDAR.1995.598994
• Jošić, H., & Žmuk, B. (2022). A machine learning approach to forecast international trade: The case of Croatia. Business Systems Research, 13(3), 144–160. https://doi.org/10.2478/bsrj-2022-0030
• Kelle, A. C., & Yüce, H. (2022). MQTT trafiğinde DoS saldırılarının makine öğrenmesi ile sınıflandırılması ve modelin SHAP ile yorumlanması. Journal of Materials and Mechatronics: A, 3(1), 50–62. https://doi.org/10.55546/jmm.995091
• Kottou, E. M., Grubelich, T. A., & Wang, X. (2020). Bilateral trade flow prediction models enhanced by wavelet and machine learning algorithms. 2020 International Conference on Computational Science and Computational Intelligence (CSCI), 1510–1516. https://doi.org/10.1109/CSCI51800.2020.00279
• Kulkarni, A., & More, A. (2016). Formulation of a prediction index with the help of WEKA tool for guiding stock market investors. Oriental Journal of Computer Science and Technology, 9, 212–225. https://doi.org/10.13005/ojcst/09.03.07
• Liaw, A., & Wiener, M. (2001). Classification and regression by randomForest. Forest, 23.
• Linnemann, H. (1966). An econometric study of international trade flows. North-Holland Pub. Co.
• Loh, W.-Y. (2011). Classification and regression trees. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 1, 14–23. https://doi.org/10.1002/widm.8
• McCallum, J. (1995). National borders matter: Canada-U.S. regional trade patterns. The American Economic Review, 85(3), 615–623. http://www.jstor.org/stable/2118191
• Melitz, J. (2008). Language and foreign trade. European Economic Review, 52(4), 667–699. https://doi.org/10.1016/j.euroecorev.2007.05.002
• Nuroğlu, E. (2012). Estimating trade flows of Turkey using panel data analysis and neural networks. Bartın Üniversitesi İİBF Dergisi, 3, 13–34.
• Pöyhönen, P. (1963). A tentative model for the volume of trade between countries. Weltwirtschaftliches Archiv, 90, 93–100. http://www.jstor.org/stable/40436776
• Quimba, F. M. A., & Barral, M. A. A. (2018). Exploring neural network models in understanding bilateral trade in APEC: A review of history and concepts. Philippine Institute for Development Studies (PIDS), Issues 2018–33. https://hdl.handle.net/10419/211053
• Quinlan, J. R. (1986). Induction of decision trees. Machine Learning, 1(1), 81–106. https://doi.org/10.1007/BF00116251
• Rokach, L., & Maimon, O. (2014). Data mining with decision trees: Theory and applications (2nd ed.). World Scientific. https://doi.org/10.1142/9097
• Sahour, H., Gholami, V., Torkman, J., Vazifedan, M., & Saeedi, S. (2021). Random forest and extreme gradient boosting algorithms for streamflow modeling using vessel features and tree-rings. Environmental Earth Sciences, 80, 1–15. https://doi.org/10.1007/s12665-021-10054-5
• Tinbergen, J. (1962). Shaping the world economy: Suggestions for an international economic policy. https://hdl.handle.net/1765/16826
• Wohl, I., & Kennedy, J. (2018). Neural network analysis of international trade. US International Trade Commission. Washington, DC, USA.