Short-term forecast of gold price using generalized autoregressive conditional heteroscedastic models

Abstract

Gold is used in many industries and it is popular as a good investment. However, its price can fluctuate widely. There are many mathematical models that can be used to forecast gold prices. In this study, the Generalised Autoregressive Conditional Heteroscedastic (GARCH) and Autoregressive Integrated Moving Average (ARIMA) models are developed to produce short term forecasts of gold prices. GARCH model is developed due to it is ability to capture the volatility by the nonconstant of conditional variance while forecasts produced by the ARIMA model are used as a benchmark. Comparison of forecasts produced by GARCH and ARIMA models are based on two performance measures: mean absolute percentage error (MAPE) and root mean square error (RMSE). In this study, analyses are done by using Minitab and E-Views software. In general, it can be concluded that the GARCH model is a potential method for forecasting trading day data of gold prices.