Future Forecasting of Turkey’s Chemical Imports: Performance Analysis of Machine Learning and Ensemble Learning Methods

Year 2025, Volume: 35 Issue: 1, 261 - 278, 24.01.2025

Kamil Abdullah Eşidir

https://doi.org/10.18069/firatsbed.1580620

Abstract

This study evaluates the performance of machine learning and ensemble learning methods to predict the future values of Turkey's chemical imports. Linear Regression, Random Forest, Rational Quadratic Regression, Support Vector Machine and XGBoost models are used. Data are obtained from reliable sources such as TurkStat and CBRT. Macroeconomic variables include Turkey's Imports, Chemicals and Chemical Products Production Index, Monthly Average Dollar Exchange Rate, Manufacturing Industry Production Index, Oil Barrel Prices and Chemicals Exports. According to the analysis results, the XGBoost model has the highest accuracy and generalization ability. The model performed well in the training, test and cross-validation sets with the lowest error rates and the highest R² values. SHAP analysis reveals that Turkey Imports and Chemicals Production Index variables have the highest impact. The projections made with the XGBoost model provide important insights into the future course of Turkey's chemical imports and are critical for economic planning and trade strategies. The model's capability facilitates strategic decisions for policymakers and the business community.

Keywords

Chemicals Imports, Machine Learning, Ensemble Learning, Economic Forecasting, XGBoost

Türkiye’nin Kimyasal Madde İthalatının Gelecek Tahmini: Makine Öğrenmesi ve Topluluk Öğrenme Yöntemleri Performans Analizi

Abstract

Bu çalışma, Türkiye'nin kimyasal madde ithalatının gelecekteki değerlerini tahmin etmek amacıyla makine öğrenmesi ve topluluk öğrenme yöntemlerinin performansını değerlendirmektedir. Doğrusal Regresyon, Rastgele Orman, Rasyonel Kuadratik Regresyon, Destek Vektör Makinesi ve XGBoost modelleri kullanılmıştır. Veriler, TÜİK ve TCMB gibi güvenilir kaynaklardan elde edilmiştir. Makroekonomik değişkenler arasında Türkiye İthalatı, Kimyasallar ve Kimyasal Ürünler Üretim Endeksi, Aylık Ortalama Dolar Kuru, İmalat Sanayi Üretim Endeksi, Petrol Varil Fiyatları ve Kimyasal Madde İhracatı yer almaktadır. Analiz sonuçlarına göre, XGBoost modeli en yüksek doğruluk ve genelleme yeteneğine sahiptir. Model, eğitim, test ve çapraz doğrulama setlerinde en düşük hata oranları ve en yüksek R² değerleri ile başarılı performans göstermiştir. SHAP analizi, Türkiye İthalatı ve Kimyasallar Üretim Endeksi değişkenlerinin en yüksek etkiye sahip olduğunu ortaya koymuştur. XGBoost modeli ile yapılan projeksiyonlar, Türkiye'nin kimyasal madde ithalatının gelecekteki seyrine dair önemli bilgiler sunmakta, ekonomik planlama ve ticari stratejiler için kritik öneme sahiptir. Modelin yeteneği, politika yapıcılar ve iş dünyası için stratejik kararları kolaylaştırmaktadır.

Keywords

Kimyasal Madde İthalatı, Makine Öğrenmesi, Topluluk Öğrenme, Ekonomik Tahmin, XGBoost

References

• Al Marzooqi, F. I. and Redouane, A. (2024). Predicting real estate prices using machine learning in abu dhabi. Iraqi Journal of Science, 1689-1706. https://doi.org/10.24996/ijs.2024.65.3.40
• Bayrak, T. (2020). A machine-learning-based model for forecasting medical device foreign trade. Eskişehir Technical University Journal of Science and Technology a - Applied Sciences and Engineering, 21(4), 477-485. https://doi.org/10.18038/estubtda.803546
• Bhagwat, A., Baets, B. D., Steen, A., Vlaeminck, B., & Fievez, V. (2012). Prediction of ruminal volatile fatty acid proportions of lactating dairy cows based on milk odd- and branched-chain fatty acid profiles: new models, better predictions. Journal of Dairy Science, 95(7), 3926-3937. https://doi.org/10.3168/jds.2011-4850
• Biau, G. and Scornet, E. (2016). A random forest guided tour. Test, 25(2), 197-227. https://doi.org/10.1007/s11749-016-0481-7
• Bilgili, M., Keiyinci, S., & Ekinci, F. (2022). One-day ahead forecasting of energy production from run-of-river hydroelectric power plants with a deep learning approach. Scientia Iranica, 0(0), 0-0. https://doi.org/10.24200/sci.2022.58636.5825
• Brabenec, T. and Šuleř, P. (2020). Machine learning forecasting of cr and prc balance of trade. SHS Web of Conferences, 73, 01004. https://doi.org/10.1051/shsconf/20207301004
• Broeck, G. V. d., Lykov, A., Schleich, M., & Suciu, D. (2022). On the tractability of shap explanations. Journal of Artificial Intelligence Research, 74, 851-886. https://doi.org/10.1613/jair.1.13283
• Broeren, M. L. M., Saygin, D., & Patel, M. K. (2014). Forecasting global developments in the basic chemical industry for environmental policy analysis. Energy Policy, 64, 273-287. https://doi.org/10.1016/j.enpol.2013.09.025
• Cao, S. and Hu, Y. (2024). Interpretable machine learning framework to predict gout associated with dietary fiber and triglyceride-glucose index. Nutrition &Amp; Metabolism, 21(1). https://doi.org/10.1186/s12986-024-00802-2
• Cao, T., She, D., Zhang, X., & Yang, Z. (2022). Understanding the influencing factors and mechanisms (land use changes and check dams) controlling changes in the soil organic carbon of typical loess watersheds in china. Land Degradation &Amp; Development, 33(16), 3150-3162. https://doi.org/10.1002/ldr.4378
• Chen, T. and Guestrin, C. (2016). Xgboost: a scalable tree boosting system.. https://doi.org/10.48550/arxiv.1603.02754
• Crescimanno, M., Galati, A., & Bal, T. (2014). The role of the economic crisis on the competitiveness of the agri-food sector in the main mediterranean countries. Agricultural Economics (Zemědělská Ekonomika), 60(2), 49-64. https://doi.org/10.17221/59/2013-agricecon
• Dutschmann, T. and Baumann, K. (2021). Evaluating high-variance leaves as uncertainty measure for random forest regression. Molecules, 26(21), 6514. https://doi.org/10.3390/molecules26216514
• Eberly, L. E. (2007). Multiple linear regression. Topics in Biostatistics, 165-187. https://doi.org/10.1007/978-1-59745-530-5_9
• Elias, R., Fang, L., & Wahab, M. (2011). Electricity load forecasting based on weather variables and seasonalities: a neural network approach. Icsssm11. https://doi.org/10.1109/icsssm.2011.5959472
• EMİRHAN, P. N. and TURGUTLU, E. (2023). İşgücü talebi̇ uluslararasi ti̇carete tepki̇ veri̇yor mu? Türk i̇malat sektöründen kanitlar. Öneri Dergisi, 18(59), 187-201. https://doi.org/10.14783/maruoneri.1075714
• Erdem, Z. B. (2010). The assessment of coal's contribution to sustainable energy development in Turkey. Energy Exploration &Amp; Exploitation, 28(2), 117-129. https://doi.org/10.1260/0144-5987.28.2.117
• Erduman, Y., Eren, O., & Gül, S. (2020). Import Content of Turkish Production And Exports: A Sectoral Analysis. Central Bank Review, 20(4), 155-168. https://doi.org/10.1016/j.cbrev.2020.07.001
• Fantke, P. and Ernstoff, A. (2017). Lca of chemicals and chemical products. Life Cycle Assessment, 783-815. https://doi.org/10.1007/978-3-319-56475-3_31
• Ferrat, L. A., Goodfellow, M., & Terry, J. R. (2018). Classifying dynamic transitions in high dimensional neural mass models: a random forest approach. PLOS Computational Biology, 14(3), e1006009. https://doi.org/10.1371/journal.pcbi.1006009
• Flyvbjerg, B. (2006). From nobel prize to project management: getting risks right. Project Management Journal, 37(3), 5-15. https://doi.org/10.1177/875697280603700302
• Ge, X., Ding, J., Jin, X., Wang, J., Chen, X., Li, X., … & Xie, B. (2021). Estimating agricultural soil moisture content through uav-based hyperspectral images in the arid region. Remote Sensing, 13(8), 1562. https://doi.org/10.3390/rs13081562
• Gür, Y. E., & Eşidir, K. A. (2024). Türkiye Hurda Demir Çelik İthalatının Gelecek Değerlerinin Derin Öğrenme, Makine Öğrenmesi ve Topluluk Öğrenme Yöntemleri ile Öngörülmesi. Alanya Akademik Bakış, 8(3), 885-908. https://doi.org/10.29023/alanyaakademik.1497646
• Gür, Y. E. (2024). Stock Price Forecasting Using Machine Learning and Deep Learning Algorithms: A Case Study for the Aviation Industry. Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 36(1), 25-34. https://doi.org/10.35234/fumbd.1357613
• Gür, Y. E. (2024). Development and application of machine learning models in US consumer price index forecasting: Analysis of a hybrid approach. Data Science in Finance and Economics, 4(4), 469-513. https://doi.org/10.3934/DSFE.2024020
• Gür, Y. E. (2024). Innovation in the dairy industry: forecasting cow cheese production with machine learning and deep learning models. International Journal of Agriculture Environment and Food Sciences, 8(2), 327-346. https://doi.org/10.31015/jaefs.2024.2.9
• Huertas‐Tato, J. and Brito, M. (2018). Using smart persistence and random forests to predict photovoltaic energy production. Energies, 12(1), 100. https://doi.org/10.3390/en12010100
• Hui, D., Wang, J., Le, X., Shen, W., & Ren, H. (2012). Influences of biotic and abiotic factors on the relationship between tree productivity and biomass in china. Forest Ecology and Management, 264, 72-80. https://doi.org/10.1016/j.foreco.2011.10.012
• Ji, H., Fu, X., & Cheng, K. (2014). Engineering drawing man-hour forecasting based on bp-ga in design of chemical equipment. 2014 20th International Conference on Automation and Computing. https://doi.org/10.1109/iconac.2014.6935487
• Jošić, H. and Žmuk, B. (2022). A machine learning approach to forecast international trade: the case of croatia. Business Systems Research Journal, 13(3), 144-160. https://doi.org/10.2478/bsrj-2022-0030
• Kishi, H. and Sekine, Y. (2003). Simple prediction of atmospheric concentration of hydrophilic compounds based on the classification of industrial uses. QSAR &Amp; Combinatorial Science, 22(3), 396-398. https://doi.org/10.1002/qsar.200390029
• Li, G., Yu, Z., Zheng, B., Qi, B., Su, Z., & Wang, D. (2023). Voltage sag source location based on the random forest. Journal of Physics: Conference Series, 2584(1), 012144. https://doi.org/10.1088/1742-6596/2584/1/012144
• Li, N., & Li, M. (2022). Forecast of Chemical Export Trade Based on PSO‐BP Neural Network Model. Journal of Mathematics, 2022(1), 1487746. https://doi.org/10.1155/2022/1487746
• Liang, W., Luo, S., Zhao, G., & Wu, H. (2020). Predicting hard rock pillar stability us ing gbdt, xgboost, and lightgbm algorithms. Mathematics, 8(5), 765. https://doi.org/10.3390/math8050765
• MacCallum, R. C., Zhang, S., Preacher, K. J., & Rucker, D. D. (2002). On the practice of dichotomization of quantitative variables. Psychological Methods, 7(1), 19-40. https://doi.org/10.1037/1082-989x.7.1.19
• Marill, K. A. (2004). Advanced statistics: linear regression, part ii: multiple linear regression. Academic Emergency Medicine, 11(1), 94-102. https://doi.org/10.1111/j.1553-2712.2004.tb01379.x
• Medeiros, M. C. (2022). Forecasting With Machine Learning Methods. Advanced Studies in Theoretical and Applied Econometrics, 111-149. https://doi.org/10.1007/978-3-031-15149-1_4
• Meng, F., Weng, K., Shallal, B., Chen, X., & Mourshed, M. (2018). Forecasting algorithms and optimization strategies for building energy management & demand response. Sp 2018. https://doi.org/10.3390/proceedings2151133
• Metlek, S., Kandilli, C., & Kayaalp, K. (2021). Prediction of the effect of temperature on electric power in photovoltaic thermal systems based on natural zeolite plates. International Journal of Energy Research, 46(5), 6370-6382. https://doi.org/10.1002/er.7575
• Mishra, S., Srivastava, R., Muhammad, A., Amit, A., Chiavazzo, E., Fasano, M., … & Asinari, P. (2023). The impact of physicochemical features of carbon electrodes on the capacitive performance of supercapacitors: a machine learning approach. Scientific Reports, 13(1). https://doi.org/10.1038/s41598-023-33524-1
• Moon, J., Ke, F., & Sokolikj, Z. (2020). Automatic assessment of cognitive and emotional states in virtual reality‐based flexibility training for four adolescents with autism. British Journal of Educational Technology, 51(5), 1766-1784. https://doi.org/10.1111/bjet.13005
• Moshiri, S. and Kheirandish, E. (2023). Global impacts of oil price shocks: the trade effect. Journal of Economic Studies, 51(1), 126-144. https://doi.org/10.1108/jes-08-2022-0455
• Nas, S., Akboz Caner, A., & Ergin Ünal, A. (2024). Türkiye’de Enflasyon Oranlarının Makine Öğrenme Yöntemi ile Tahmini. Gaziantep Üniversitesi Sosyal Bilimler Dergisi, 23(3), 1029-1045. https://doi.org/10.21547/jss.1371005
• Naseem, M., Chaudhary, K., Sharma, B. N., & Lal, A. G. (2019). Using ensemble decision tree model to predict student dropout in computing science. 2019 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE). https://doi.org/10.1109/csde48274.2019.9162389
• Özemre, M. and Kabadurmuş, Ö. (2020). A big data analytics based methodology for strategic decision making. Journal of Enterprise Information Management, 33(6), 1467-1490. https://doi.org/10.1108/jeim-08-2019-0222
• Palkovits, R. and Palkovits, S. (2019). Using artificial intelligence to forecast water oxidation catalysts. ACS Catalysis, 9(9), 8383-8387. https://doi.org/10.1021/acscatal.9b01985
• Pisuttinusart, C., Jatuporn, C., Suvanvihok, V., & Seerasarn, N. (2022). Forecasting the import demand for chemical fertilizer in Thailand. The EUrASEANs: journal on global socio-economic dynamics, (3 (34)), 61-70. https://doi.org/10.35678/2539-5645.3(34).2022.61-70
• Polat, K., Şentürk, Ü. K., & Arıcan, M. (2023). A novel cuffless blood pressure prediction: uncovering new features and new hybrid ml models. Diagnostics, 13(7), 1278. https://doi.org/10.3390/diagnostics13071278
• Renas Rajab Asaad and M. Abdulazeez, A. (2024). Comprehensive classification of iris flower species: a machine learning approach. Indonesian Journal of Computer Science, 13(1). https://doi.org/10.33022/ijcs.v13i1.3717
• Sercu, P. and Uppal, R. (2003). Exchange rate volatility and international trade: a general-equilibrium analysis. European Economic Review, 47(3), 429-441. https://doi.org/10.1016/s0014-2921(01)00175-1
• Sevim, A., Demir, İ., Höfte, M., Humber, R. A., & Demirbağ, Z. (2009). Isolation and characterization of entomopathogenic fungi from hazelnut-growing region of Turkey. BioControl, 55(2), 279-297. https://doi.org/10.1007/s10526-009-9235-8
• Svetnik, V., Liaw, A., Tong, C., Culberson, J., Sheridan, R. P., & Feuston, B. P. (2003). Random forest: a classification and regression tool for compound classification and qsar modeling. Journal of Chemical Information and Computer Sciences, 43(6), 1947-1958. https://doi.org/10.1021/ci034160g
• Şener, S., Savrul, M., & Aydın, O. (2014). Structure of small and medium-sized enterprises in turkey and global competitiveness strategies. Procedia - Social and Behavioral Sciences, 150, 212-221. https://doi.org/10.1016/j.sbspro.2014.09.119
• Şimşek, A. I., Bulut, E., Gur, Y. E., & Tarla, E. G. (2024). A novel approach to Predict WTI crude spot oil price: LSTM-based feature extraction with Xgboost Regressor. Energy, 309, 133102. https://doi.org/10.1016/j.energy.2024.133102
• Şimşek, A. I. (2024). Forecasting consumer price index using macroeconomic variables: a comparative analysis of machine learning and deep learning approaches. Bingöl Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, (28), 15-29. https://doi.org/10.29029/busbed.1394983
• Truong, N., Ngo, N., & Pham, A. (2021). Forecasting time-series energy data in buildings using an additive artificial intelligence model for improving energy efficiency. Computational Intelligence and Neuroscience, 2021, 1-12. https://doi.org/10.1155/2021/6028573
• Türkiye Cumhuriyet Merkez Bankası Elektronik Veri Dağıtım Sistemi (TCMB-EVDS), https://evds2.tcmb.gov.tr/index.php?/evds/serieMarket, , Erişim Tarihi: 22.05.2024.
• Türkiye İstatistik Kurumu (2024), www.tuik.gov.tr, Erişim Tarihi: 15.05.2024.
• Tyass, I., Khalili, T., Rafik, M., Bellat, A., Raihani, A., & Mansouri, K. (2023). Wind speed prediction based on statistical and deep learning models. International Journal of Renewable Energy Development, 12(2), 288-299. https://doi.org/10.14710/ijred.2023.48672
• Walker, J. D., Dimitrov, S., & Mekenyan, O. G. (2003). Using hpv chemical data to develop qsars for non‐hpv chemicals: opportunities to promote more efficient use of chemical testing resources. QSAR &Amp; Combinatorial Science, 22(3), 386-395. https://doi.org/10.1002/qsar.200390028
• Wang, L., Wang, X., Chen, A., Jin, X., & Che, H. (2020). Prediction of type 2 diabetes risk and its effect evaluation based on the xgboost model. Healthcare, 8(3), 247. https://doi.org/10.3390/healthcare8030247
• Wang, N., Liu, W., Sun, S., & Wang, Q. (2021). The influence of complexity of imported products on total factor productivity. Mobile Information Systems, 2021, 1-7. https://doi.org/10.1155/2021/3384068
• Wang, Y. (2022). Import and export trade forecasting algorithm based on blockchain security and pso optimized hybrid rvm model. Security and Privacy, 6(2). https://doi.org/10.1002/spy2.218
• Wen, B., Li, R., Zhao, X., Ren, S., Chang, Y., Zhang, K., … & Zhu, X. (2021). A quadratic regression model to quantify plantation soil factors that affect tea quality. Agriculture, 11(12), 1225. https://doi.org/10.3390/agriculture11121225
• Wen, J., Zhang, Y., Yang, G., He, Z., & Zhang, W. (2019). Path loss prediction based on machine learning methods for aircraft cabin environments. IEEE Access, 7, 159251-159261. https://doi.org/10.1109/access.2019.2950634
• Wie, Y. M., Lee, K. G., Lee, K. H., Ko, T. S., & Lee, K. H. (2020). The experimental process design of artificial lightweight aggregates using an orthogonal array table and analysis by machine learning. Materials, 13(23), 5570. https://doi.org/10.3390/ma13235570
• Yap, M., Johnston, R. L., Foley, H., MacDonald, S., Kondrashova, O., Tran, K., … & Waddell, N. (2021). Verifying explainability of a deep learning tissue classifier trained on rna-seq data. Scientific Reports, 11(1). https://doi.org/10.1038/s41598-021-81773-9
• Zareipour, H., Cañizares, C. A., & Bhattacharya, K. (2010). Economic impact of electricity market price forecasting errors: a demand-side analysis. IEEE Transactions on Power Systems, 25(1), 254-262. https://doi.org/10.1109/tpwrs.2009.2030380
• Zhang, Z., Xin, Q., & Li, W. (2021). Machine learning‐based modeling of vegetation leaf area index and gross primary productivity across north america and comparison with a process‐based model. Journal of Advances in Modeling Earth Systems, 13(10). https://doi.org/10.1029/2021ms002802
• Zhu, C., Liu, X., & Chen, D. (2024). Prediction of digital transformation of manufacturing industry based on interpretable machine learning. Plos One, 19(3), e0299147. https://doi.org/10.1371/journal.pone.0299147
• Zolfigol, M. A., Azizian, S., Torabi, M., Yarie, M., & Notash, B. (2024). The importance of nonstoichiometric ratio of reactants in organic synthesis. Journal of Chemical Education, 101(3), 877-881. https://doi.org/10.1021/acs.jchemed.3c00530