Gold and Silver Price Prediction using Hybrid Machine Learning Models

Gold and silver are two of the most precious metals in great demand due to their industrial, electrical, and decorative uses. They have gained high popularity in terms of investments and trading. The prediction of their price with good accuracy can be beneficial for people and investors. This paper attempts to evaluate the potency of machine learning models in anticipating the future price of two precious metals (gold and silver) in the Indian market. In particular, the implementation of hybrid machine learning models like CNN (Convolutional Neural Network) and CNN-RNN (CNN with Recurrent Neural Network) have been analyzed through this work. The experiments have been done using the daily traded data from January 2021 to August 2022. The accuracy of the models has been measured using Mean Absolute Percentage Error (MAPE). From the results, it is found that the RNN model has performed well only in the case of gold price prediction, whereas other models have achieved nearly acceptable accuracy.