Is there a connection between AI, Machine Learning and soybeans? Oh, but there is!

Another project where AI and Machine Learning play a crucial role in making it all go next level, only this time for small and medium-size soybeans futures contracts traders and trading organizations. 

After winning a contract from the European Space Agency for the project SPATIAL (Soybeans Price forecAsting based on saTellite-derIved services and Artificial intelligence), Hypertech is now developing a proof-of-concept (PoC) prototype for forecasting soybeans futures contracts price moves using Artificial Intelligence models based on financial & macroeconomic features and Earth Observation products. The project already started in October 2021 and is expected to finish in October 2022. 

The solution for soybeans traders 

SPATIAL is addressing the small and medium size soybeans futures contracts traders and trading organizations, which are currently underserved. This is due to the fact that competing services or means to acquire soybeans futures contacts price forecasts are currently not affordable for this market segment, and many times require special financial knowledge or in-house analytic skills and infrastructures. Future contracts, which is an agreement to buy or sell an asset at a future date at an agreed-upon price, is a very popular asset class because it has standardised terms and is traded on an exchange, where prices are settled on a daily basis until the end of the contract. Companies and individuals can trade futures contracts either for hedging or speculative reasons. 

Focus on the potential 

The demand and consequently the production of soy is expected to rise in the following years, positively affecting the whole value chain, namely farmers, buyers, animal feed producers, biofuel producers, as well as food producers. Consequently, the volume and the value of the soy commodity exchange market (including soybeans, soy meal and soy oil) are also expected to expand, creating important opportunities for traders, brokers and consulting companies. Therefore, the need for systems supporting soy purchasing/trading decisions is evident.

In this context, SPATIAL addresses a niche developing market with obvious potential and targets the bulk of the users/potential customers involved in the global soy market value chain. Soybeans and other commodities as financial assets, can be traded by anyone through many trading products. Agricultural commodities futures contracts, as a hedging method, is very important to the relative companies, organisations and farmers that produce, purchase or do business with soybeans products through these contracts. They can buy or sell products in the future against any risk since the price and the contract expiration are pre-agreed. 

Analysing the solution

SPATIAL ambition is to democratise access to state-of-the-art forecasting technology that builds on proprietary ML models and algorithms for processing multiple data sources. The Artificial Intelligence models and the Earth Observation products can help all the above users to make more efficient, more reliable, and more profitable decisions through the quantified price estimations of the soybeans futures contracts that the SPATIAL project will provide.

Through the project, SPATIAL will bring to life two distinct Machine Learning (ML) models. One is for soybeans crop yield forecasting and the other for prediction of soybeans futures contracts price moves. This is to demonstrate the feasibility of the method, the benefits of integrating Copernicus EO products, and to showcase the potential of such an approach. The solution, in general, is a multi-parametric and multi-feature AI system that will ultimately be offered as a service, at a reasonable cost for soybeans commodities purchasing, trading, and SME companies helping them fortify multi-million soybeans purchasing decisions against price risks. 

What SPATIAL will offer

SPATIAL interface will present the prediction, the prediction result, and a series of economic and other metadata, so that the user can appreciate the quality and the success of the prediction, providing at the same time, useful information regarding the economic conditions at the time of the reference date. Finally, the system will present an overview of how the prediction was calculated, giving the contribution of each feature to support an Explainable AI paradigm. 

All the above are based on the ideal combination of expertise of Hypertech and NOA. On one hand, Hypertech will bring its ‘mastery’ in financial assets price forecasting through machine learning and predictive analytics models for financial asset prices prediction, where all is based on traditional and alternative data sources along with our multi-year expertise in financial markets dynamics and deep knowledge on the key factors affecting commodities prices. On the other hand, everything will be addressed with the NOA’s highly prospered skills in the development and deployment of Space-based applications for estimating soybeans crop yields and production.