Estimating a garch model for gold price Returns: A Bayesian Approach

The frequency of rerun of any stock consistently fluctuates due to the competing forces of supply and demand responding to changes in the share prices by investors. Historically, gold prices have generally produced favourable returns during both challenging and thriving times, positioning gold as a means for safeguarding and enhancing wealth. In this study, we present a Bayesian Generalized Auto-Regressive Conditional Heteroskedastic (GARCH) volatility model for daily gold price returns based on the most recent 2500 daily prices. The present research aims to showcase the use of the stan-garch function from the ‘bayesforecast’ of (R-package) to fit a GARCH (1, 1) model to the gold price returns data, assuming Student-t and normal error distributions. The results of the research show that the model effectively captures the data. It is also concluded that the effects of prior shocks will result in a lasting impact on the future volatility of the daily gold price returns.