Türkiye'de İhracat-İthalat Oranını Etkileyen Dinamikler: Ekonometri ve Makine Öğrenmesi Yaklaşımıyla Hibrit Model Önerisi

Öz

Dış ticaretle ilgili göstergeler, geleneksel olarak bir para birimi veya ülkenin gayri safi yurtiçi hasılası’nın oranı olarak ölçülmektedir. Diğer yandan, ihracatın ithalatı karşılama oranı birimsiz bir gösterge olarak, ekonomilerin hem zaman içindeki hem de diğer ülkelerle dış ticaret performanslarını karşılaştırırken daha faydalı sonuçlar vermektedir. Bu çalışmada, bu oranı etkileyen makro ekonomik ve finansal belirleyiciler hem ekonometrik olarak hem de makine öğrenmesi yöntemi kullanılarak incelenmiştir. Bu kapsamda ARDL yöntemi, ilk olarak 2010-2021 yılları arasında normalize GSYİH, döviz kuru, TÜFE, ÜFE, ham petrol ile Türkiye'nin REI oranı arasındaki ilişkiyi aylık olarak araştırmak için kullanılmıştır. Uzun vadeli analiz, TL'nin ABD doları karşısında %1'lik değer kaybının REI oranını 0,7 puan artırdığını göstermiştir. Ek olarak, TÜFE'deki %1'lik bir artış REI'yi 1,9 puan artırırken, ÜFE'deki %1'lik bir artış REI'de -0.8 puanlık bir düşüşe neden olmaktadır. Daha sonra, değişkenler arasındaki örüntü, bir makine öğrenmesi yöntemi olan ikinci dereceden destek vektör makineleri (SVM) ile analiz edilmiştir. Son olarak, yeni ARDL-SVM hibrit yöntemi geliştirilmiş ve değişkenler arasındaki örüntü incelenmiştir. Bulgular, ekonometrik yöntemin değişkenler arasındaki ilişkileri yorumlamada daha geniş bir perspektif sunmasına rağmen, geliştirilen ARDL-SVM yönteminin değişkenler arasındaki örüntüleri daha başarılı şekilde yakaladığını ortaya koymuştur.

Anahtar Kelimeler

• Uluslararası ticaret
• İhracatın ithalata oranı
• ARDL
• SVM
• Türkiye

The Dynamics Affecting the Export-Import Ratio in Turkey: A Hybrid Model Proposal with Econometrics and Machine Learning Approach

Öz

The indicators related to foreign trade are conventionally measured in a currency or as the ratio of the country's gross domestic products. The ratio of exports to imports, alternatively, provides more useful results when comparing the foreign trade performance of economies both over time and with other countries as a unit-free indicator. In this study, the macroeconomics and financial determinants affecting this ratio are examined both econometrically and using the machine learning method. In this context, the autoregressive distributed lag model method was first used to investigate the relationship between normalized gross domestic products, exchange rate, consumer price index, producer price index, crude oil and Turkey's ratio of exports to imports rate between 2010- 2021, monthly. Long-term analysis showed that the 1% depreciation of the Turkish Liras against the US dollar increased the ratio of exports to imports rate by 0.7 points. In addition, a 1% increase in consumer price index will increase ratio of exports to imports by 1.9 points, while a 1% increase in producer price index will cause a -0.8 point decrease on the ratio of exports to imports. Then, the pattern between the variables was analyzed with quadratic support vector machine, a machine learning method. Finally, the novel ARDL-SVM hybrid method was developed, and the pattern between the variables was examined. The findings revealed that although the econometric method provided a broader scope for interpreting the relationships between variables, the developed ARDL- SVM method successfully captured patterns between variables.

Anahtar Kelimeler

• International trade
• Export to import ratio
• ARDL
• SVM
• Turkey

Kaynakça

1. Ahmed, Z., Zhang, B., & Cary, M. (2021). Linking economic globalization, economic growth, financial development, and ecological footprint: Evidence from symmetric and asymmetric ARDL. Ecological Indicators, 121, 107060. https://doi.org/https://doi.org/10.1016/j.ecolind.2020.107060
2. Altıntaş, H. (2013). Türkiye’de petrol fiyatları, ihracat ve reel döviz kuru ilişkisi: ARDL Sınır testi yaklaşımı ve dinamik nedensellik analizi. International Journal, 9(19), 1–30. https://doi.org/10.11122/ijmeb.2013.9.19.459
3. Bakshi, S. S., Jaiswal, R. K., & Jaiswal, R. (2021). Efficiency check using cointegration and machine learning approach: Crude oil futures markets. Procedia Computer Science, 191, 304–311. https://doi.org/https://doi. org/10.1016/j.procs.2021.07.038
4. Bardi, W., & Hfaiedh, M. A. (2021). International trade and economic growth: Evidence from a panel ARDL-PMG approach. International Economics and Economic Policy, 18(4), 847–868. https://doi.org/10.1007/s10368- 021-00507-4
5. Cooke, D. (2010). Openness and inflation. Journal of Money, Credit and Banking, 42(2–3), 267–287. https://doi. org/https://doi.org/10.1111/j.1538-4616.2009.00287.x
6. Cortes, C., & Vapnik, V. (1995). Support-Vector networks. Machine Learning, 20(3), 273–297. https://doi. org/10.1023/A:1022627411411
7. Engle, R. F., & Granger, C. W. J. (1987). Co-Integration and Error Correction: Representation, estimation, and testing. Econometrica, 55(2), 251–276. https://doi.org/10.2307/1913236
8. Granger, C. W. J., & Newbold, P. (1974). Spurious regressions in econometrics. Journal of Econometrics, 2(2), 111–120. https://doi.org/https://doi.org/10.1016/0304-4076(74)90034-7

9. Guleryuz, D. (2022). Estimation of soil temperatures with machine learning algorithms—Giresun and Bayburt stations in Turkey. Theoretical and Applied Climatology, 147(1), 109–125. https://doi.org/10.1007/s00704- 021-03819-2
10. Guleryuz, D., & Ozden, E. (2020). The prediction of brent crude oil trend using LSTM and Facebook prophet.
11. European Journal of Science and Technology, 20, 1–9. https://doi.org/10.31590/ejosat.759302
12. Güneş, Ş., & Konur, F. (2013). Türkiye ekonomisinde dışa açıklık ve enflasyon ilişkisi üzerine ampirik bir analiz.
13. Eskişehir Osmangazi Üniversitesi İktisadi ve İdari Bilimler Dergisi, 8(2), 7–20.
14. Johansen, S. (1988). Statistical analysis of cointegration vectors. Journal of Economic Dynamics and Control, 12(2), 231–254. https://doi.org/https://doi.org/10.1016/0165-1889(88)90041-3
15. Johansen, S., & Juselius, K. (2009). Maximum likelihood estimation and inference on cointegration with applications to the demand for money. Oxford Bulletin of Economics and Statistics, 52(2), 169–210. https://doi. org/10.1111/j.1468-0084.1990.mp52002003.x
16. Lin, H.Y. (2011). Openness and inflation revisited. International Research Journal of Finance and Economics, 37, 40–45.
17. Martinez, R., & Iyer, V. (2013). Openness and inflation: evidence from nine Eastern European nations. International Business & Economics Research Journal (IBER), 13(1), 21. https://doi.org/10.19030/iber.v13i1.8353
18. Mikic, M., & Gilbert, J. (2009), Trade Statistics in Policymaking - A Handbook of Commonly Used Trade Indices And Indicators (Revised Edition), United Nations Economic and Social Commission for Asia and the Pacific (ESCAP)
19. Morley, B. (2006). Causality between economic growth and immigration: An ARDL bounds testing approach.
20. Economics Letters, 90(1), 72–76. https://doi.org/https://doi.org/10.1016/j.econlet.2005.07.008
21. Mukhtar, S., Adamu, B., Ibrahim Abdullahi, S., Shehu, K., & Buba, S. (2022). International trade and export dynamics in Nigeria: An ARDL vector error correction analysis. Journal of Humanities, Arts and Social Science, 6, 50–61. https://doi.org/10.26855/jhass.2022.01.005
22. Narayan, P. K. (2005). The saving and investment nexus for China: Evidence from cointegration tests. Applied Economics, 37(17), 1979–1990. https://econpapers.repec.org/RePEc:taf:applec:v:37:y:2005:i:17:p:1979-1990
23. Narayan, P. K., & Smyth, R. (2006). What determines migration flows from low-income to high-income countries? An empirical investigation of Fiji-US migration 1972-2001. Contemporary Economic Policy, 24(2), 332–342. https://doi.org/10.1093/cep/byj019
24. Odhiambo, N. M. (2009). Energy consumption and economic growth nexus in Tanzania: An ARDL bounds testing approach. Energy Policy, 37(2), 617–622. https://doi.org/https://doi.org/10.1016/j.enpol.2008.09.077
25. OECD. (2022). Main Economic Indicators. https://doi.org/10.1787/data-00052-en
26. Ozden, E., & Guleryuz, D. (2021). Optimized machine learning algorithms for investigating the relationship between economic development and human capital. Computational Economics. https://doi.org/10.1007/ s10614-021-10194-7
27. Pesaran, M. H., & Shin, Y. (1997). An autoregressive distributed-lag modelling approach to cointegration analysis. In Econometrics and Economic Theory in the 20th Century: The Ragnar Frisch Centennial Symposium (pp. 371–413). Cambridge University Press. https://doi.org/10.1017/CCOL0521633230.011
28. Pesaran, M. H., Shin, Y., & Smith, R. J. (2001). Bounds testing approaches to the analysis of level relationships.
29. Journal of Applied Econometrics, 16(3), 289–326. https://doi.org/10.1002/jae.616
30. Raghuramapatruni, R., & V. S.C. (2020). An appraisal of the impact of international trade on economic growth of India- through the ARDL approach. International Journal of Economics and Business Administration, VIII(Issue 2), 376–387. https://doi.org/10.35808/ijeba/468
31. Rüping, S. (2001). SVM kernels for time series analysis (Issue 2001,43). Universität Dortmund, Sonderforschungsbereich 475 - Komplexitätsreduktion in Multivariaten Datenstrukturen. http://hdl.handle. net/10419/77140
32. Sun, Y., Zhang, X., & Wang, S. (2020). A hierarchical forecasting model for China’s Foreign trade. Journal of
33. Systems Science and Complexity, 33(3), 743–759. https://doi.org/10.1007/s11424-020-8070-y
34. Thao, D. T., & Jian Hua, Z. (2016). ARDL bounds testing approach to cointegration: Relationship international trade policy reform and foreign trade in Vietnam. International Journal of Economics and Finance, 8(8), 84. https://doi.org/10.5539/ijef.v8n8p84
35. Thomas, C. (2012). Trade openness and inflation: Panel data evidence for the Caribbean. International Business & Economics Research Journal (IBER), 11(5), 507. https://doi.org/10.19030/iber.v11i5.6969
36. Vita, G., & Abbott, A. (2004). Real exchange rate volatility and US Exports: An ARDL bounds testing approach. Economic Issues, 9.
37. Waliullah, Kakar, M., Kakar, R., & Khan, W. (2010). The determinants of Pakistan’s trade balance: An ARDL cointegration approach. The Lahore Journal of Economics, 15. https://doi.org/10.35536/lje.2010.v15.i1.a1
38. Wang, Y., Li, J., Gu, J., Zhou, Z., & Wang, Z. (2015). Artificial neural networks for infectious diarrhea prediction using meteorological factors in Shanghai (China). Applied Soft Computing, 35, 280–290. https://doi.org/https://doi.org/10.1016/j.asoc.2015.05.047
39. Worldbank. (2022). Trade of the World’s GDP. Worldbank Open Data. https://data.worldbank.org/indicator/ NE.TRD.GNFS.ZS
40. Wu, C.F., Huang, S.C., Chang, T., Chiou, C.C., & Hsueh, H.P. (2020). The nexus of financial development and economic growth across major Asian economies: Evidence from bootstrap ARDL testing and machine learning approach. Journal of Computational and Applied Mathematics, 372, 112660. https://doi.org/10.1016/j. cam.2019.112660
41. Yu, L., Wang, S., & Lai, K. K. (2008). Forecasting China’s foreign trade volume with a kernel-based hybrid econometric-ai ensemble learning approach. Journal of Systems Science and Complexity, 21(1), 1–19. https:// doi.org/10.1007/s11424-008-9062-5