ARCHIVES

Abstract: Gold price prediction holds significant importance for investors, economists, and policy makers due to its role as a global safe-haven asset and its influence on financial markets. In this paper, we propose a machine learning (ML) approach for predicting gold prices based on historical data and relevant economic indicators. The dataset includes various features such as gold prices, United State Oil ETF (USO Value), S&P 500 Index (S&P500), Euro against USD pair (EUR/USD), Shares Silver Trust (SLV). We preprocess the data to handle missing values, normalize features, and split it into training and testing sets. We experiment with different ML algorithms, including to Support Vector Machines, K-Neighbors Classifier, Random Forests Classifier, Gaussian NB, Decision Tree Classifier and Random Forests Regressor. We tune hyperparameters using techniques like grid search or randomized search to optimize model performance. Evaluation metrics such as mean absolute error (MAE), mean squared error (MSE), and R-squared are used to assess the accuracy and robustness of the models. We compare the performance of different algorithms and feature combinations to identify the most effective approach for gold price prediction. The results demonstrate the effectiveness of ML algorithms in predicting gold prices, with certain models outperforming traditional econometric approaches. Additionally, feature importance analysis provides insights into the key drivers of gold price movements. Overall, this paper contributes to the growing body of research on forecasting financial markets using machine learning techniques and provides valuable insights for investors and stakeholders in the gold market.