Forecasting Foreign Trade in Türkiye’s Agri-Food Sector: A Comparative Analysis of SARIMA, LSTM, and GRU Models

Year 2025, Volume: 12 Issue: 4, 1007 - 1026, 17.10.2025

Zekiye Şengül Abedı

https://doi.org/10.30910/turkjans.1756810

Abstract

In a global economic conjuncture shaped by geopolitical tensions, climate shocks, and supply chain vulnerabilities, the accurate forecasting of foreign trade data underpins proactive policy design for national food security and economic stability. This study systematically compares the forecasting performance of Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) models, two modern deep learning approaches, and the traditional Seasonal ARIMA (SARIMA) model to forecast Türkiye's agriculture, food, and beverage foreign trade series. Monthly data for the period 2004-2020 were used to train the models, and their forecasting performance was measured using MAE, MSE, RMSE, and MAPE metrics over the test period 2021-2025. The results of the analysis indicate that the most appropriate model depends on the structure of the series. Whereas the SARIMA model outperforms in export series with regular seasonal patterns, the GRU model performs better in import series with complex nonlinear dynamics. These results underline the need to adopt a data-driven and problem-specific modeling strategy rather than relying on a universal "best model". In light of these findings, this study recommends the adoption of a hybrid "model portfolio" approach that combines SARIMA models for predictable seasonal trends and deep learning methods for managing volatile and shock-prone environments to enhance the resilience of national trade strategies.

Keywords

Foreign trade forecasting , deep learning , LSTM , GRU , SARIMA , model comparison , Turkish agricultural sector

References

• Adanacioglu, H., & Yercan, M. (2012). An analysis of tomato prices at wholesale level in Turkey: An application of SARIMA model. Custos e Agronegocio, 8(4), 52–75.
• Airlangga, G. (2024). Advanced forecasting of maize production using SARIMAX models: An analytical approach. JURNAL MEDIA INFORMATIKA BUDIDARMA, 8(1), 361–370. https://doi.org/10.30865/mib.v8i1.7268
• Aker, Y. (2022). Analysis of price volatility in BIST 100 index with time series: Comparison of Fbprophet and LSTM model. European Journal of Science and Technology, 35, 89–93. https://doi.org/10.31590/ejosat.1066722 Aksoy, M. A., & Beghin, J. C. (2005). Global agricultural trade and developing countries. World Bank Publications. Alexiadis, S. (2017). Forecasting agricultural production using co-integration analysis. Land Use Policy, 61, 466–474. https://doi.org/10.1016/j.landusepol.2016.11.038
• Arlt, J., Trcka, P., & Arltová, M. (2017). The problem of the SARIMA model selection for the forecasting purpose. Statistika: Statistics and Economy Journal, 97(4), 25–32.
• ArunKumar, K. E., Kalaga, D. V., Mohan Sai Kumar, Ch., Kawaji, M., & Brenza, T. M. (2022). Comparative analysis of Gated Recurrent Units (GRU), long Short-Term memory (LSTM) cells, autoregressive Integrated moving average (ARIMA), seasonal autoregressive Integrated moving average (SARIMA) for forecasting COVID-19 trends. Alexandria Engineering Journal, 61(10), 7585–7603. https://doi.org/10.1016/j.aej.2022.01.011
• Bagheri, A., Mohammadi Peyhani, H., & Akbari, M. (2014). Financial forecasting using ANFIS networks with Quantum-behaved Particle Swarm Optimization. Expert Systems with Applications, 41(14), 6235–6250. https://doi.org/10.1016/j.eswa.2014.04.003
• Bars, T., Uçum, İ., & Akbay, C. (2018). ARIMA modeli ile Türkiye fındık üretim projeksiyonu. Kahramanmaraş Sütçü İmam Üniversitesi Tarım ve Doğa Dergisi, 21, 154–160. https://doi.org/10.18016/ksutarimdoga.v21i41625.473029
• Başaran Caner, C. B., & Engindeniz, S. (2020). Türkiye’de pamuk üretiminin ARIMA modeli ile tahmini. Tarım Ekonomisi Dergisi, 26(1), 63–70. https://doi.org/10.24181/tarekoder.681079
• Çakan, V. A. (2020). Türkiye yaş incir üretimi ve kuru incir ihracatı için öngörü: ARIMA modeli yaklaşımı. Tekirdağ Ziraat Fakültesi Dergisi, 17(3), Article 3. https://doi.org/10.33462/jotaf.684893
• Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014). Learning phrase representations using RNN encoder-decoder for statistical machine translation (arXiv:1406.1078). arXiv. https://doi.org/10.48550/arXiv.1406.1078
• Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014). Empirical evaluation of gated recurrent neural networks on sequence modeling (arXiv:1412.3555). arXiv. https://doi.org/10.48550/arXiv.1412.3555
• Çukur, T., & Çukur, F. (2021). ARIMA modeli ile Türkiye bal üretim öngörüsü. Tarım Ekonomisi Araştırmaları Dergisi, 7(1), 31–39.
• Doğar, D., Çiçek, A., & Ayyıldız, M. (2024). Türkiye’de süt ve süt ürünleri üretiminin mevsimsel ARIMA modeli ile tahmini. Turkish Journal of Agricultural and Natural Sciences, 11(3), 642–653. https://doi.org/10.30910/turkjans.1389143
• Eroğlu, N. A., Bozoğlu, M., Topuz, B. K., & Başer, U. (2019). Forecasting the amount of beef production in Turkey. Tarım Ekonomisi Araştırmaları Dergisi, 5(2), 101–107.
• FAO. (2009). Global agriculture towards 2050. https://www.fao.org/fileadmin/templates/wsfs/docs/Issues_papers/HLEF2050_Global_Agriculture.pdf
• FAO. (2024). The state of agricultural commodity markets 2024. Trade and nutrition: Policy coherence for healthy diets. FAO. https://doi.org/10.4060/cd2144en
• Gao, S., Alawad, M., Schaefferkoetter, N., Penberthy, L., Wu, X.-C., Durbin, E. B., Coyle, L., Ramanathan, A., & Tourassi, G. (2020). Using case-level context to classify cancer pathology reports. PLoS ONE, 15(5), e0232840. https://doi.org/10.1371/journal.pone.0232840
• Göksu, E., & Saner, G. (2021). Çam balı üretici satış fiyatlarının Box-Jenkins modeli ile öngörüsü. Turkish Journal of Forestry, 22(2), 111–116. https://doi.org/10.18182/tjf.917303
• Gopinath, M., Batarseh, F. A., Beckman, J., Kulkarni, A., & Jeong, S. (2021). International agricultural trade forecasting using machine learning. Data & Policy, 3, e1. https://doi.org/10.1017/dap.2020.22
• Greff, K., Srivastava, R. K., Koutník, J., Steunebrink, B. R., & Schmidhuber, J. (2017). LSTM: a search space odyssey | IEEE Journals & Magazine | IEEE Xplore. IEEE Transactions on Neural Networks and Learning Systems, 28(10), 2222–2232. https://doi.org/10.1109/TNNLS.2016.2582924
• Güler, D., Saner, G., & Naseri, Z. (2017). Yağlı tohumlu bitkiler ithalat miktarlarının ARIMA ve yapay sinir ağları yöntemleriyle tahmini. Balkan and Near Eastern Journal of Social Sciences, 03(01), 60–70.
• Gür, Y. E. (2024). Forecasting the euro exchange rate using deep learning algorithms and machine learning algorithms. İstanbul Ticaret Üniversitesi Sosyal Bilimler Dergisi, 23(49), 1435–1456. https://doi.org/10.46928/iticusbe.1379268
• Hawkes, C., & Murphy, S. (2010). An overview of global food trade. In Trade, Food, Diet and Health: Perspectives and Policy Options. John Wiley & Sons.
• İnan, C., & Oktay, E. (2023). Fourier serileri ile modifiye edilmiş modelleri kullanarak Borsa İstanbul (BİST) üzerine bir uygulama. Turkish Research Journal of Academic Social Science, 5(2), 81–91.
• Kadakoğlu, B., & Karlı, B. (2022). Türkiye ekonomisinde tarıma dayalı sanayinin önemi ve gelişim süreci. Turkish Journal of Science and Engineering, 4(1), 50–59. https://doi.org/10.55979/tjse.1107524
• Kaplan, K., & Çiçek, A. (2025). Estimation of Turkey hazelnut export quantity and prices with ARIMA model. Journal of Agricultural Faculty of Gaziosmanpaşa University, 42(1), 27–35. https://doi.org/10.55507/gopzfd.1629321
• Karadağ, K. (2022). Hibrit derin öğrenme modelleri ile hisse senedi fiyat tahmini. [Yüksek Lisans Tezi]. Trakya Üniversitesi Sosyal Bilimler Enstitüsü.
• Karakaş, G. (2021). Forecasting of natural honey yield in Turkey through ARIMA model. Journal of Agricultural Faculty of Gaziosmanpaşa University, 38(3), 166–172. https://doi.org/10.13002/jafag4777
• Kurnadipare, A. I., Amaliya, S., Notodiputro, K. A., Angraini, Y., & Mualifah, L. N. A. (2025). Comparing forecasts of agricultural sector export values using SARIMA and long short-term memory models. BAREKENG: Journal of Mathematics and Its Applications, 19(1), 385–396. https://doi.org/10.30598/barekengvol19iss1pp0385-0396
• Li, J., Li, G., Liu, M., Zhu, X., & Wei, L. (2022). A novel text-based framework for forecasting agricultural futures using massive online news headlines. International Journal of Forecasting, 38(1), 35–50. https://doi.org/10.1016/j.ijforecast.2020.02.002
• Li, S. (2024). Inflation forecasting using a hybrid LSTM-SARIMA model based on discrete wavelet transform. 73, 100–108. https://doi.org/10.54254/2754-1169/73/20231419
• Liu, R., Jiang, Y., & Lin, J. (2022). Forecasting the volatility of specifc risk for stocks with LSTM. Procedia Computer Science, 202, 111–114. https://doi.org/10.1016/j.procs.2022.04.015
• Madaan, L., Sharma, A., Khandelwal, P., Goel, S., Singla, P., & Seth, A. (2019). Price forecasting & anomaly detection for agricultural commodities in India. Proceedings of the 2nd ACM SIGCAS Conference on Computing and Sustainable Societies, 52–64. https://doi.org/10.1145/3314344.3332488
• Majka, M. (2024). Seasonal time series analysis: Why SARIMA outshines ARIMA. https://www.researchgate.net/publication/384196885_Seasonal_Time_Series_Analysis_Why_SARIMA_Outshines_ARIMA
• Milovanović, M., Radosavac, A., & Knežević, D. (2018). State of agro-food foreign trade in Bosnia and Herzegovina. Economics of Agriculture, 65(3), 1059–1070. https://doi.org/10.5937/ekoPolj1803059M
• Nayak, G. H. H., Alam, M. W., Avinash, G., Singh, K. N., Ray, M., & Kumar, R. R. (2024). N-BEATS deep learning architecture for agricultural commodity price forecasting. Potato Research. https://doi.org/10.1007/s11540-024-09789-y
• Noa-Yarasca, E., Osorio Leyton, J. M., & Angerer, J. P. (2024). Deep learning model effectiveness in Füforecasting limited-size aboveground vegetation biomass time series: Kenyan grasslands case study. Agronomy, 14(2), 349. https://doi.org/10.3390/agronomy14020349
• Okere, E. E., & Balyan, V. (2025). A deep learning-based prediction and forecasting of tomato prices for the cape town fresh produce market: A model comparative analysis. Forecasting, 7(2), 19. https://doi.org/10.3390/forecast7020019
• Ou, S.-L. (2012). Forecasting agricultural output with an improved grey forecasting model based on the genetic algorithm. Computers and Electronics in Agriculture, 85, 33–39. https://doi.org/10.1016/j.compag.2012.03.007
• Özbek, F. Ş., & Fidan, H. (2009). Estimation of pesticides usage in the agricultural sector in Turkey using artificial neural network (ANN). Journal of Animal & Plant Sciences, 4(3), 373–378.
• Öztürk, G. (2022). Forecasting pistachio production in Turkey: A comparison of ARIMA, Grey, and Exponential Smoothing models. Philippine Agricultural Scientist, 105(2), 180–186.
• Panicker, N. K. K., & Valarmathi, J. (2024). A hybrid SARIMA-LSTM approach for improved time series prediction of aerosol optical depth across Delhi, India. Journal of Theoretical and Applied Information Technology, 102(11).
• Paul, R. K., & Garai, S. (2022). Wavelets based artificial neural network technique for forecasting agricultural prices. Journal of the Indian Society for Probability and Statistics, 23(1), 47–61. https://doi.org/10.1007/s41096-022-00128-3
• Polina, P., Ganesan, S., Karunarathne, L., & Somasiri, N. (2024). Time series analysis for tractor sales using SARIMAX and deep learning models. International Journal of Computer Communication and Informatics, 6(1), 27–57. https://doi.org/10.34256/ijcci2413
• Ramos, K. G., & Ativo, I. J. O. (2023). Forecasting monthly prices of selected agricultural commodities in the Philippines using ARIMA model. International Journal of Research Publication and Reviews, 4(1), 1983–1993.
• Ray, S., Lama, A., Mishra, P., Biswas, T., Sankar Das, S., & Gurung, B. (2023). An ARIMA-LSTM model for predicting volatile agricultural price series with random forest technique. Applied Soft Computing, 149, 110939. https://doi.org/10.1016/j.asoc.2023.110939
• Republic of Türkiye Ministry of Trade. (2025). Genel tarım sektörü (p. 11). İhracat Genel Müdürlüğü Tarım Ürünleri Daire Başkanlığı. https://ticaret.gov.tr/data/5b8700a513b8761450e18d81/Genel%20Tar%C4%B1m%20Sekt%C3%B6r%C3%BC%20Raporu.pdf
• Rodikov, G., & Antulov-Fantulin, N. (2022). Can LSTM outperform volatility-econometric models? (arXiv:2202.11581). arXiv. https://doi.org/10.48550/arXiv.2202.11581
• Saxena, R., Singh, R., Agarwal, P., Kumar, R., & Raman, M. S. (2023). Structure, performance and competitiveness in Indian agricultural exports. In S. A. Narula & S. P. Raj (Eds.), Sustainable Food Value Chain Development (pp. 295–317). Springer Nature. https://doi.org/10.1007/978-981-19-6454-1_14
• Soy Temür, A. (2019). İşletmelerin satış bütçelerinin oluşturulmasında ARIMA, LSTM ve hibrit modellerin karşılaştırılması: Üretim işletmesi örneği. [Doktora tezi]. Sakarya Üniversitesi İşletme Enstitüsü.
• Sumner, D. A., Hallstrom, D. G., & Lee, H. (1998). Trade policy and the effects of climate forecasts on agricultural markets. American Journal of Agricultural Economics, 80(5), 1102–1108. https://doi.org/10.2307/1244212
• Tayib, S. A. M., Nor, S. R. M., & Norrulashikin, S. M. (2021). Forecasting on the crude palm oil production in Malaysia using SARIMA Model. Journal of Physics: Conference Series, 1988(1), 012106. https://doi.org/10.1088/1742-6596/1988/1/012106 TGDF. (2025). TGDF dış ticaret bülteni – Aralık 2024. https://www.tgdf.org.tr/tgdf-dis-ticaret-bulteni-aralik-2024/
• Türk, M. M. (2024). Türkiye ekonomisinde reel efektif döviz kuru, reel ihracat ve reel ithalat arasındaki ilişkinin asimetrik nedensellik testleri ile analizi (2013-2023). Cumhuriyet Üniversitesi İktisadi ve İdari Bilimler Dergisi, 25(1), 96–111. https://doi.org/10.37880/cumuiibf.1332134
• TurkStat. (2025). Türkiye İstatistik Kurumu. https://www.tuik.gov.tr/ Accessed 20 June 2025
• Turmuş, E., & Güneş, E. (2025). Ankara ilinde bitkisel ürün işleyen ve ihracatında bulunan gıda sanayi işletmelerinin yapısı ve sorunları üzerine bir araştırma. Çukurova Tarım ve Gıda Bilimleri Dergisi, 40(1), 21–34.
• Uçar, K., Güler, D., & Engindeniz, S. (2021). Türkiye’de kayısı üretiminin ARIMA modeli ile tahmini. Tarım Ekonomisi Dergisi, 27(2), 55–62. https://doi.org/10.24181/tarekoder.941416
• UNCTAD. (2024). Trade against hunger: Exploring trade actions to fight acute food insecurity and the threat of famine. United Nations. https://doi.org/10.18356/9789211070545
• Wang, J., Wang, Z., Li, X., & Zhou, H. (2022). Artificial bee colony-based combination approach to forecasting agricultural commodity prices. International Journal of Forecasting, 38(1), 21–34. https://doi.org/10.1016/j.ijforecast.2019.08.006
• WTO. (2024). Members discuss farm policies, address food security, technology transfer, transparency. https://www.wto.org/english/news_e/news24_e/agri_26sep24_e.htm
• Yamak, P. T., Yujian, L., & Gadosey, P. K. (2020). A comparison between ARIMA, LSTM, and GRU for time series forecasting. Proceedings of the 2019 2nd International Conference on Algorithms, Computing and Artificial Intelligence, 7p. https://doi.org/10.1145/3377713.3377722
• Yıldız, M., & Atış, E. (2019). Türkiye organik incir ihraç fiyatının Arma yöntemi ile tahmini. Tarım Ekonomisi Dergisi, 25(2), 141–147. https://doi.org/10.24181/tarekoder.583972
• Zarzycki, K., & Ławryńczuk, M. (2021). LSTM and GRU neural networks as models of dynamical processes used in predictive control: A Comparison of models developed for two chemical reactors. Sensors, 21(16), 5625. https://doi.org/10.3390/s21165625