Advanced Machine Learning Techniques for Predicting Gold and Silver Futures

Abstract

This research focuses on predicting the future values of gold and silver futures by employing advanced machine learning algorithms. Traditional econometric models often struggle with commodity prices’ non-linear and dynamic nature. To address this, the study evaluates the performance of four unconventional machine learning algorithms: Gaussian Processes, Quantile Regression Forests, Extreme Learning Machines, and Support Vector Regression with an RBF kernel. The dataset used includes monthly trading data for gold and silver futures. The research findings indicate that these machine- learning models significantly enhance prediction accuracy. Support Vector Regression with an RBF kernel demonstrated the highest accuracy for gold futures predictions, while Extreme Learning Machines performed competitively for silver futures. This study highlights the potential of advanced machine learning techniques in financial forecasting, providing valuable insights for traders and investors in optimizing their strategies.