Prediction of sunflower production parameters in Serbia using the Arima model

Nedeljković, Miroslav and Petrović, Jana and Dokić, Dragan (2026) Prediction of sunflower production parameters in Serbia using the Arima model. Western Balkan Journal of Agricultural Economics and Rural Development (WBJAERD), 8 (1). pp. 37-48. ISSN 2683-4693

Text Nedeljković, Petrović, Dokić - (WBJAERD), 8 (1) 2026.pdf - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (423kB)

Abstract

The main goal of research is analysis and forecast of sunflower production in Serbia, using a time series model, aiming to identify trends and provide accurate forecasts for the short-term period. Research is especially focused on assessment of the potential of using statistical models to predict the production trends of one of the most important oilseeds in Serbia. The paper uses the Autoregressive Integrated Moving Average (ARIMA) model, according to the Box-Jenkins methodological approach. Modeling process has included testing the stationarity of time series, determining the optimal specification of model based on the ACF and PACF functions, estimating the model parameters, as well as checking its adequacy through the use of standard statistical measures of forecast precision. Derived research results suggest that the used model provides a reliable short-term forecast of sunflower production, identifying the key trends in analyzed time frame. Forecasts indicate relatively stable changes in production, with specific oscillations that are characteristic of agricultural production, while depend on the agro-economic and climatic situation. Research results have important practical significance for the organization of agricultural production, regulation of the oilseed market and decision-making within agrarian policy. Based on performed analysis, it can be pointed out that ARIMA models are efficient analytical tool for short-term production forecasting, while additional assessment of economic and climatic variables is advised for future research in order to improve the accuracy of predictions.

Actions (login required)

View Item