Estimation of Turkey hazelnut export quantity and prices with ARIMA model

• Published in: Journal of Agricultural Faculty of Gaziosmanpasa University Vol. 42; no. 1; pp. 27 - 35
• Main Authors: Kaplan, Kaan, Çiçek, Adnan
• Format: Journal Article
• Language: English
• Published: 30.04.2025
• DOI: 10.55507/gopzfd.1629321
• ISSN: 1300-2910

Hazelnut is of strategic importance for Turkey and is a product subject to international trade. Turkey, which realizes 64% of the world hazelnut production, is also the country that exports the most. It is important to estimate the future hazelnut price, export amount and income from exports to maintain the country’s position. Hazelnut export unit price, hazelnut export quantity and hazelnut export value variables in Turkey between 1961 and 2023 were used and forecasted with ARIMA model for 2024, 2025 and 2026. Statistical error evaluation criteria such as mean absolute percentage error (MAPE), mean absolute error (MAE), root mean square error (RMSE), normalized Bayesian information criterion (BIC) etc. were used to test the validity of the ARIMA model, which indicated that the model was reliable. In addition, the augmented Dickey-Fuller test (ADF) and Phillips-Perron test (PP) unit root tests were applied to determine the stationarity levels of the series. The series was stationarity at different levels and the Ljung-Box significance levels of the series were appropriate for the models. It is predicted that export unit price and hazelnut export value will follow an increasing trend in the next three years, while hazelnut export quantity will follow a fluctuating course. In addition, it can be said that the export quantity will continue to follow a fluctuating course over the years with the effect of periodicity in production, while the export unit price and export value will continue its upward trend in a fluctuating manner with the effect of the crisis experienced after 2005.