Comparative Analysis of ARIMA and GARCH Models for Forecasting Spot Gold Prices and Their Volatility: A Time Series Study

Gold's reputation as a secure investment offering protection against inflation has solidified its status as a preferred asset among investors. As a result, forecasting gold prices while accounting for their inherent patterns and trends is of significant interest to the research community. A notable trend in existing gold price forecasting studies is their concentration on US Dollar-denominated gold prices, thereby overlooking potential influences from exchange rates and tax rates specific to gold prices denominated in Rupees. This study aims to establish an autoregressive integrated moving average (ARIMA) model as a foundational benchmark for gold price forecasting, followed by the development of a generalized autoregressive conditional heteroskedastic (GARCH) model for the purpose of comparative analysis. Both the ARIMA and GARCH methods are commonly utilized by researchers for time-series forecasting and have been explored independently in various studies. We considered gold prices over a duration of ten years, covering the period from March 2013 to March 2023. Through meticulous utilization of the EViews software, this research rigorously evaluates the outcomes produced by both models. The findings underscore the comparable effectiveness of both ARIMA and GARCH models in adeptly capturing and predicting the intricate dynamics of gold price movements. The research findings could provide valuable insights to investors seeking to enhance their investment strategies in the gold market.