Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/329237 
Year of Publication: 
2024
Citation: 
[Journal:] Economies [ISSN:] 2227-7099 [Volume:] 12 [Issue:] 11 [Article No.:] 310 [Year:] 2024 [Pages:] 1-24
Publisher: 
MDPI, Basel
Abstract: 
Soybeans, a vital source of protein for animal feed and an essential industrial raw material, are the most traded agricultural commodity worldwide. Accurate price forecasting is crucial for maintaining a resilient global food supply chain and has significant implications for agricultural economics and policymaking. This review examines over 100 soybean price forecast models published in the last decade, evaluating them based on the specific markets they target - futures or spot - while highlighting how differences between these markets influence critical model design decisions. The models are also classified into AI-powered and traditional categories, with an initial aim to conduct a statistical analysis comparing the performance of these two groups. This process unveiled a fundamental gap in best practices, particularly regarding the use of common benchmarks and standardised performance metrics, which limits the ability to make meaningful cross-study comparisons. Finally, this study underscores another important research gap: the lack of models forecasting soybean futures prices in Brazil, the world's largest producer and exporter. These insights provide valuable guidance for researchers, market participants, and policymakers in agricultural economics.
Subjects: 
soybean
price forecast
artificial intelligence
futures market
spot market
Persistent Identifier of the first edition: 
Creative Commons License: 
cc-by Logo
Document Type: 
Article

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.