Base Metals Price Forecasting From Alternate Data Sources Using AI
Implementation Time:
9 months
Solution Provider: AI Singapore
Founded in 2008, Four Elements Capital is a specialised commodity asset manager focused on delivering absolute returns through a fundamental systematic investment approach.
Based in Singapore, Four Elements Capital is research driven with a strong dedication to processes & analysis.
- Base metals, used as materials for the construction of infrastructures and varlous types fundamental for human development
- Base metal prices are stochastic and highly non-linear as Infuenced by many factors with complex relationships
- Thev are also affected by macro-economic situations such as currency exchange rates and government policy changes e.g.increase in import tax rate
- Valuable information can be mined from alternative data sources such as news, specialised forums, reports, social media and demand forecasts by various courtiers and related companies
How can Four Elements Capital leverage machine learning techniques to improve the prediction of spot prices of base metals using market data (trading prices and volumes); macroeconomic data; supply and demand data; third party estimations and relevant social media information (alternative data)?
A machine learning framework consisting of three components:
- Classification of price movement direction using an ensemble of multiple machine learning and deep learning models
- Regression of price predictions using a deep learning model
- Filter mechanism to align the outputs from the above
News and Analyst Report Indicators were also developed to extract intelligence from alternative data sources and calibrate the classification outputs
Outcomes
- Machine learning framework outperformed traditional linear model benchmarks on out-of-sample forecast accuracy for 1, 3, 5 and 10 day horizons
- New technologies were developed to utilise alternative data relevant to base metal trading e.g. news and analysts reports to derive short-term price movement indicators
- Source code has been deployed and tested on Four Elements server
- Implementing the AI solution in the financial world would strengthen and develop Singapore as a leading Fintech hub
- Potential spin-off and scaling of this technology would help Singapore financial institutions, business owners as well as government entities better manage their risk exposure
PDF Document
This is the last content of this tab. If you do not see any resources above, it means the solution provider have not provided any resources. Feel free to contact Solution Provider for more information.
Implementation Time
9 months
Use Case Brochure
