Agric. Econ. - Czech, 2026, 72(7):461-472 | DOI: 10.17221/102/2025-AGRICECON

Corn price forecasts in the shadow of the Russian–Ukrainian conflictOriginal Paper

László Vancsura ORCID..., Arnold Csonka ORCID...

Department of Marketing and Supply Chain Analysis, Institute of Agricultural and Food Economics, Hungarian University of Agriculture and Life Sciences, Kaposvár, Hungary

The conflict between Russia and Ukraine has caused serious disruption to agricultural markets, affecting both food prices and food security. In our study, we examine how global economic problems, such as COVID-19 or wartime conditions, affect corn prices and their predictability. We used deep neural network models [Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU)], experimenting with both simple and hybrid versions, as well as univariate and multivariate types. The results show that the GRU model outperformed the other models in predicting corn prices. It was also found that multivariate models tended to yield more accurate results than their univariate counterparts, suggesting that forecast quality can be significantly improved by including additional variables. In the robustness analysis, it was shown that the COVID-19 crisis in 2020 and the Russia–Ukraine war in 2022 led to a deterioration in model performance. The research contributes to the development of forecasting methods for the corn market and provides a basis for decision-making strategies for market players. To ensure methodological rigour, the forecasting results were statistically validated using the Diebold–Mariano test.

Keywords: deep learning; forecasting; market shocks; predictive modelling; time series analysis

Received: March 6, 2025; Revised: February 24, 2026; Accepted: February 27, 2026; Published: July 31, 2026  Show citation