Multivariate Machine Learning Models for Gold Price Forecasting

TITLE	Multivariate Machine Learning Models for Gold Price Forecasting
ABSTRACT	Gold, being a world known safe-haven investment has very volatile price behavior that is affected by various economics indicator results and market dynamics. Proper prediction in the gold prices is a necessity to the investors, policy makers and financial institutions. This paper is a machine learning approach to regressing previous gold prices based on historical data and macro economic indicators including the S&P 500 (SPX), silver prices (SLV), oil prices (USO) and EUR/USD exchange rates. Multiple regression models and benchmarks such as: Linear Regression, Decision Tree, Random Forest, Support Vector Regressor (SVR), XGBoost, and Long Short-Term Memory (LSTM) neural networks are used and compared in the project. Preprocessing of data, feature engineering, and hyperparameter optimization were done to make the model more effective. The models were validated using evaluation measures like Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R 2 Score. The outcomes show that ensemble methods such as Random Forest and deep learning such as LSTM perform more accurately as compared to traditional when used in the prediction. The results show that time-series analytics combined with machine learning are effective to create reliable and scalable forecasts of gold prices.
AUTHOR	Abhishek Dodwad, Adarsh Ballur, Agastya T, Gomathi Thiyagarajan Department of MCA, CMR Institute of Technology, Bengaluru, India
PUBLICATION DATE	2025-09-05
VOLUME	12
DOI	DOI:10.15680/IJARETY.2025.1204065
PDF	65_Multivariate Machine Learning Models for Gold Price Forecasting.pdf
KEYWORDS

Article

Email

[email protected]