Ensuring food security is a global challenge, particularly in low-income countries where food prices affect access to nutritious food. The instability of global agricultural commodity (AC) prices exacerbates food insecurity, with international trade restrictions and market disruptions further complicating the situation. Although online platforms exist for monitoring food prices, there is still a need for accessible, detailed forecasts for non-specialists. This paper proposes the Agricultural Commodity Analysis and Forecasts (AGRICAF) methodology, integrating explainable machine learning (XML) and econometric techniques to analyse and forecast global ACs prices up to one year ahead across different horizons. This innovative integration allows us to model complex interactions while providing clear, interpretable results. We demonstrate the utilization of AGRICAF, applying it to three major commodities and explaining how different factors impact prices across months and forecast horizons. By facilitating access to reliable forecasts of AC prices, AGRICAF can advance a fairer and sustainable food system.
