Gold Price Prediction using Machine Learning and Deep Learning Algorithms

Today, different markets and economic sectors are directly or indirectly affected by gold price; thus, its prediction is a big challenge for both investors and researchers. On the other hand, the nonstationary and nonlinear patterns of gold price data cause the prediction process even more complex. To address this challenge, a hybrid model was developed in this paper to predict gold price, with a concentration on enhancing accuracy through considering the gold price data characteristics. To do this, Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and Gated recurrent units (GRU) were used to deal with the nonstationary and nonlinear nature of the gold price data. The former was first applied to the decomposition of time-series data of gold price into a number of components. Then, GRU was applied to the prediction of the components. To end with, all the components’ prediction results were summed up to attain the final prediction result. The efficiency of the developed model was evaluated using real-world gold data, which confirmed its superiority over the standard methods used for comparison.

Keywords:

Deep learning , Decomposition , Forecasting , Gold Price