Gold Price Prediction Using Hybrid Deep Learning by Integrating LSTM-ANN Network with GARCH Model

Abstract

Gold investment is increasingly favored by the public as a relatively stable long-term investment instrument. However, the unpredictable fluctuations in gold prices make it difficult for investors to make appropriate investment decisions. Complex factors such as market volatility, economic news, changes in monetary policy, inflation, and geopolitical uncertainty lead to sharp movements in gold prices, which are difficult to predict using conventional methods. This study aims to develop an accurate gold price prediction model using a hybrid deep learning approach by integrating the LSTM-ANN Network and the GARCH model. This hybrid method combines the strengths of the LSTM-ANN Network in capturing temporal patterns and non-linear trends in historical price data, with the ability of the GARCH model to handle gold price volatility. This approach is expected to provide more accurate predictions compared to conventional forecasting methods. This study uses historical gold price data as the basis for prediction, focusing on gold price forecasting over a specific time period. The results of this study are expected to contribute to the development of commodity price prediction models, particularly gold, and provide a tool to help investors make more informed investment decisions.