International agricultural trade forecasting using machine learning

Focusing on seven major agricultural commodities with a long history of trade, this study employs data-driven analytics to decipher patterns of trade, namely using supervised machine learning (ML), as well as neural networks. The supervised ML and neural network techniques arc trained on data until 2010 and 2014, respectively. Results show the high relevance of ML models to forecasting trade patterns in near- and long-term relative to traditional approaches, which arc often subjective assessments or time-series projections. While supervised ML techniques quantified key economic factors underlying agricultural trade flows, neural network approaches provide better fits over the long term.

Automated summary

The study focused on seven major agricultural commodities and utilized data-driven analytics to decipher trade patterns, with supervised machine learning and neural networks showing higher relevance in forecasting trade patterns compared to traditional methods. While supervised machine learning quantified economic factors underlying agricultural trade flows, neural network approaches provided better fits in the long term.