A Gold Price Prediction based on the US Dollar Index using the RF-MLP Meta Model Stacking Regressor

Gold is frequently employed as a hedge in circumstances characterized by elevated inflation rates, economic recession, and other factors that give rise to economic uncertainty. The US Dollar Index is one factor that influences the price of gold. It is an index number that reflects and measures the strength of the US Dollar against six other major world currencies. Although only six currencies are listed, the US dollar index measurement process actually compares the US dollar to the currencies of 24 countries (19 of which are members of the Eurozone). The US dollar index may be employed as a benchmark for gauging the general strength of the US dollar. This research proposes an innovative approach to accurately predict gold prices based on the US dollar index, fundamental news data, and technical data. The Meta Model Stacking Regressor and Random Forest, in conjunction with the Multilayer Perceptron, yielded the most optimal results when employed in isolation. When evaluated over the three-period study, the Mean Squared Error (MSE) values for the Random Forest and Multilayer Perceptron stand-alone models were 3.536, 5.474, and 6.222, respectively. These values were superior to those observed when the Random Forest and Multilayer Perceptron were employed as standalone models.