Prediction of Gold prices in India: Performance Analysis of Various Machine Learning Models

Gold is a valuable commodity and serves as an investment opportunity for businessmen, traders and even for a family. Prior knowledge of the upcoming prices of gold will help investors make better trading decisions. This information will help traders gauge their options and know the best time to purchase gold to make sound investments to reap the maximum return for their future. This study attempts to make prediction of the gold price using Machine Learning Techniques. There are four Machine Learning models which were employed to analyze the data and they are: 1) an Artificial Neural Network (ANN) model 2) an Artificial Neural Network with a Particle Swarm Optimization Algorithm (ANN-PSO) hybrid model 3) Recurrent Neural Network (RNN) and 4) a hybrid model of Convolutional Neural Network with Recurrent Neural Network (CNN-RNN). We have used daily price data over the last four years from 1 January 2018 to 18 March 2022. It was found that the hybrid ANN-PSO model outperforms the other three models.