RBM neural networks for gold price prediction

• Main Author: Lee, Wei Ming
• Other Authors: Wang Lipo
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/167597
• Institution: Nanyang Technological University

Gold price forecasting is important due to its innate value. Its price movements are of interest to the scientific and financial communities, as a resource or an investment. Mathematical and Statistical methods along with Artificial Intelligence techniques have been applied to predicting gold prices. However, there is a gap in research on the effectiveness of Restricted Boltzmann Machine in predicting gold prices. The established Machine Learning technique is used in many other areas of financial forecasting including but not limited to stock prices, interest rate and credit risk analysis. This research proposes using Restricted Boltzmann Machine to predict gold prices. The study will then be compared to other Machine Learning techniques such as Long Short-Term Memory, Generalized Autoregressive Conditional Heteroskedasticity, Neural Networks, and Autoregressive integrated Moving Average to understand the effectiveness of the Restricted Boltzmann Machine in gold price prediction. The findings of the Restricted Boltzmann Machine’s performance metrics have outperformed most Statistical Methods but is generally weaker than other Artificial Intelligence methods employed in other studies, understood through comparison.