Integration Of Garch Models And External Factors In Gold Price Volatility Prediction: Analysis And Comparison Of Garch-M Approach

Abstract

This study investigates the volatility of gold prices by applying the Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model and extending it with the GARCH-M model, incorporating the Federal Reserve's interest rate as an external variable. The GARCH(1,1) model revealed a positive average daily return for gold, with high sensitivity to recent price changes, indicated by the significant estimation of mu and a high alpha1 value. The persistence of past volatility on current volatility is reflected by a beta1 value close to one. In the GARCH-M model development, a significant negative relationship was found between the Federal Reserve's interest rates and gold returns, suggesting that an increase in the Federal Reserve's interest rates could potentially decrease gold returns. An increase in the Log Likelihood value and improvements in information criteria such as the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) indicate that the GARCH-M model provides a better fit than the GARCH(1,1) model that uses only gold price data. The study concludes that macroeconomic factors like the Federal Reserve's interest rates play a crucial role in influencing gold price volatility, and these findings can aid investors and portfolio managers in devising more effective risk management strategies. Additionally, the findings contribute to financial theory by highlighting the importance of multivariate models in the analysis of asset price volatility.