Improving the accuracy of spot price predictions for base metals

The collaboration with Four Elements Capital on the "Base Metals Price Forecasting from Alternate Sources Using AI" has successfully completed. In this project, we developed a price forecasting framework and deployed the framework to the Alphien platform. As a result, our ML framework outperformed traditional linear model benchmarks in its accuracy for out-of-sample forecasts for one, three, five, and 10-day horizons.

Led by Dr Fuli Feng, our team also developed new indicators to extract intelligence from alternative data sources relevant to base metal trading, such as news and analyst reports, to derive short-term price movement indicators.

This project is funded under the AISG 100 Experiments project. We would like to thank our collaborator, Four Elements Capital, and AISG for their tremendous support given to our project.

Technical Papers published under this project:

- Enhancing Stock Movement Prediction with Adversarial Training

- Graph Adversarial Training: Dynamically Regularizing Based on Graph Structure

Other Papers:

- Temporal Relational Ranking for Stock Prediction

AISG 100E Project Report

Recent Posts

See All