Forecasting gold price using particle swarm optimisation and genetic algorithm based artificial neural networks

The price prediction of gold plays an important role since it is considered to be one of the most prioritised commodities in investments. The investors consider gold as a hedgerow against the unforeseen event taking place in the stock market which may lead to chaos. This research paper uses the genetic algorithm and particle swarm optimisation (PSO) based artificial neural network (ANN) models to predict the gold prices of the market. A case study has been done using the data from Perth Mint of Australia, the official bullion of the country. The performance evaluation has been done using the parameters such as correlation coefficient (R), mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE). The results obtained from the experiments suggest that while both models perform well for the gold price prediction, PSO based ANN delivers better performance as compared to GA based ANNs.