Machine Learning-Based Prediction of Gold Prices Using Economic Indicators

In furtherance to this, this research work focuses on the role of applying machine learning techniques in predicting the price of Gold using price data along with three macroeconomic variables. Thirteen models were tested in order to determine which of the three models: They also established that among the seven algorithms, namely linear regression, random forest regressor, and gradient boosting regressor, is the most useful in predicting gold price. Gradient Boosting Regressor was the optimized of all to tremendous accuracy with the help of the used algorithm that let it learn complex and non-linear features in the data inputs. They proved that the recent month and year price data, long term moving average and US Dollar Index (DXY) are the variables which affect the gold prices. The results confirm the necessity to include a broad set of economic variables in the model and the information can be reputable and valuable to investors and policy makers. Nevertheless, it could be proposed that future work would try to incorporate more features and even try to apply deep learning as an effort to improve the accuracy of the envisioned predictions.