Türkiye's Egg Export to Iraq: Performance Comparison of Seasonal ARIMA and Artificial Neural Network Models

Yıl 2024, Cilt: 10 Sayı: 2, 169 - 185

Diyar Muadh Khalil Cuma Akbay

https://doi.org/10.61513/tead.1530553

Öz

This study aims to identify the most effective model for predicting the monthly export volumes of eggs from Türkiye to Iraq by comparing two primary forecasting methods: the Seasonal Autoregressive Integrated Moving Average (SARIMA) model and the Artificial Neural Network (ANN) model. Both models were applied to monthly export data of egg products from 2010 to 2020, sourced from reliable databases such as the UN Comtrade and Turkish Statistical Institute (TURKSTAT). The performance of both models was assessed using key statistical metrics, including the Akaike Information Criterion (AIC), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and the coefficient of determination (R²). According to the results, the Feed-Forward Neural Networks (FFNN) model demonstrated superior predictive accuracy compared to the SARIMA model. This conclusion is supported by the FFNN model’s lower MAE, RMSE, and AIC values, indicating fewer forecasting errors and a better overall fit to the data. Therefore, the study concludes that the FFNN model is more effective and accurate than the SARIMA in predicting the export values of eggs from Türkiye to Iraq.

Anahtar Kelimeler

Forecasting, Artificial neural networks, Time series, Autoregressive integrated moving average, Egg export

Türkiye'nin Irak'a Yumurta İhracatı: Mevsimsel ARIMA ve Yapay Sinir Ağı Modellerinin Performans Karşılaştırması

Öz

Bu çalışma, Türkiye'den Irak'a ihraç edilen aylık yumurta miktarını tahmin etmek için en iyi modeli belirlemek amacıyla iki temel tahmin yöntemini karşılaştırmaktadır. Birinci yöntem Mevsimsel Otoregresif Bütünleşik Hareketli Ortalama (SARIMA) modeli, ikinci yöntem ise Yapay Sinir Ağı (ANN) modelidir. Her iki model de BM Comtrade ve Türkiye İstatistik Kurumu (TÜİK) resmi internet sitelerinden alınan 2010-2020 yılları arasındaki yumurta ürünleri aylık ihracat verilerine uygulanmıştır. Analiz üç yazılım programı kullanılarak gerçekleştirildi: Alyuda NeuroIntelligence, RStudio ve SPSS. Modeller AIC, MAE, RMSE ve R² metrikleri kullanılarak karşılaştırıldı. Sonuçlar, İleri Beslemeli Sinir Ağları (FFNN) modelinin SARIMA modelinden daha iyi performans gösterdiğini göstermektedir. Spesifik olarak, FFNN modeli daha az hata sergiler ve daha düşük MAE, RMSE ve AIC değerleriyle kanıtlandığı gibi, önemli ölçüde daha iyi uyum iyiliğini göstermektedir. Sonuç olarak, FFNN modelinin Türkiye'den Irak'a yumurta ihracat değerlerini tahmin etmede SARIMA modelinden daha doğru sonuçlar verdiği saptanmıştır.

Anahtar Kelimeler

Tahmin, Yapay sinir ağları, Zaman serileri, Otoregresif hareketli ortalama, Yumurta ihracatı

Kaynakça

• Abdoli, G. (2020). Comparing the prediction accuracy of LSTM and ARIMA models for time-series with permanent fluctuation. Journal of the Center for Studies and Research on Gender and Law Centre for Legal Sciences-Federal University of Paraíba, 9(2), 314-319.
• Abhinandithe, K., Madhu B., Balasubramanian S., Sahana C. (2021). A Review on the Comparison of Box- Jenkins ARIMA and LSTM of Deep Learning. International Journal of Trend in Scientific Research and Development 5(3), 409-414.
• Abraham, E.R., Mendes dos Reis, J.G., Vendrametto, O., Oliveira Costa Neto, P.L.D., Carlo Toloi, R., Souza, A.E.D., Oliveira Morais, M.D. (2020). Time Series Prediction with Artificial Neural Networks: An Analysis Using Brazilian Soybean production. Agriculture, 10(10), 475.
• Adhikari, R., Agrawal, R.K. (2013). An introductory study on time series modeling and forecasting. Lambert Academic Publishing, Germany, ISBN: 9783659335082, 67p.
• Akaike, H. (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19(6), 716-723.
• Box, G. E. P., Jenkins, G. M. (1976). Time Series Analysis: Forecasting and Control. San Francisco: Holden-Day.
• Bozkurt, Ö.Ö., Biricik, G., Tayşi, Z.C. (2017). Artificial neural network and SARIMA based models for power load forecasting in Turkish electricity market. PloS one, 12(4), 1-24.
• Fiesler, E., Beale, R. (2020). Handbook of Neural Computation. CRS Press, New York. P. 436.
• Flaherty, J., Lombardo, R. (2000). Modelling private new housing starts in Australia. In A paper presented in the Pacific-Rim Real Estate Society conference. p. 24-27.
• Groot, C., Würtz, D., (1991). Analysis of univariate time series with connectionist nets: A case study of two classical examples. Neurocomputing, 3(4), 177-192.
• Gurney, K. (2018). An introduction to neural networks. CRC press, ISBN: 9781857285031, 248p.
• Hamilton, J. D. (1994). Time Series Analysis. Princeton University Press
• Haykin, S. (1999). Neural Networks: A Comprehensive Foundation. Prentice Hall.
• Hornik, K., Stinchcombe, M., White, H. (1989). Multilayer feedforward networks are universal approximators. Neural Networks, 2(5), 359-366.
• Hyndman, R. J., Athanasopoulos, G. (2018). Forecasting: Principles and Practice. https://otexts.com/fpp3/.
• Khalil, D.M., 2022. A Comparison of Feed Forward Neural Network Models and Time Series Models for Forecasting Türkiye's Monthly Dairy Exports to Iraq. Polytechnic Journal of Humanities and Social Sciences, 3(2), pp.253-262.
• Krautmann, A., Hadley, L. (2017). Demand issues: The product market for professional sports. In Handbook of sports economics research, ISBN: 9781315093178, 288p.
• Makridakis, S., Hibon, M. (2000). The M3-Competition: results, conclusions, and implications. International Journal of Forecasting, 16(4), 451-476.
• Mehlig, B. (2019). Artificial neural networks. University of Gothenburg, arXiv e-prints, ISBN: 10-1017-9781108860604, 241p.
• Mishra, N., Soni, H.K., Sharma, S., Upadhyay, A.K. (2018). Development and Analysis of Artificial Neural Network Models for Rainfall Prediction by Using Time-Series Data. International Journal of Intelligent Systems & Applications, 10(1), 16-23.
• Pai, P. F., Lin, C. S. (2005). A hybrid ARIMA and support vector machines model in stock price forecasting. Omega, 33(6), 497-505.
• Ripley, B. D. (1996). Pattern Recognition and Neural Networks. Cambridge University Press.
• TÜİK 2023. Foreign Trade statistics. Turkish Statistical Institution.
• UN Comtrade. (2023). International Trade Statistics Database.
• Urrutia, J.D., Alano, E.D., Aninipot, P.M.R., Gumapac, K.A., Quinto, J.Q. (2014). Modeling and forecasting foreign trade of the Philippines using time series SARIMA model. European Academic Research, 11(8), 11206-11246.
• Wang, Y., Li, S., Wang, J. (2019). A hybrid model of ARIMA and ANN for price prediction in agricultural products: The case of China. Agricultural Economics (Czech Republic), 65(2), 87-98.
• Wang, Y., Li, Y., Song, Y., Rong, X. (2020). The influence of the activation function in a convolution neural network model of facial expression recognition. Applied Sciences, 10(5), 1897.
• Zhang, G., Patuwo, B.E., Hu, M.Y. (1998). Forecasting with artificial neural networks: The state of the art. International journal of forecasting, 14(1), 35-62.
• Zhang, G. P. (2003). Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing, 50, 159-175.