Comparison of ARIMA model and Artificial Neural Network in forecasting gold price

Abstract

Developing an accurate model of gold price is crucial as gold price have a great effect on the investment decisions of individuals, corporations and countries. The purpose of this study is to compare the performance of model Autoregressive Integrated Moving Average (ARIMA) and Artificial Neural Network (ANN) in gold price forecasting based on the value of root mean squared error (RMSE). Daily gold price data collected from World Gold Council dated from 3 September 2018 to 30 October 2020 is used in this study. ARIMA (4,1,0) is chosen as the best model for the time series model based on Akaike Information Criterion (AIC). Long short-term memory (LSTM) has been chosen as artificial neural network’s method to forecast the gold price. After comparing multiple step forecasting and one step ahead forecasting using ARIMA and LSTM, it is found that LSTM has smaller RMSE as compared to ARIMA. The result in this paper show that the ANN model outperforms ARIMA model in forecasting gold price.