Volatility spillover effects among gold, crude oil and US dollar index: an analysis using EGARCH Model

Abstract

Purpose: The purpose of this study is to empirically test the level of interdependence between three major assets: Gold, crude oil and USD index in terms of the return volatility spillover effects. This helps investors managing their portfolios and provides them with sense of direction on how to use these assets in portfolio diversification leading to a risk reduction benefits. Design/methodology/approach: Gold, oil and USD index volatilities are estimated using EGARCH (1, 1) model after detecting the stationarity and heteroskedasticity of the sample using the following diagnostic tests: Augmented Dickey-Fuller and ARCH test respectively. Moreover, from the estimation of the volatilities a Simultaneous Equation Model (SEM) is formulated in order to determine the volatility spillover effects through the examination of Two and Three-Stage Least Squares models. Findings: A bidirectional volatility spillover effect in terms of returns is detected from both oil and gold markets. In addition, a bidirectional volatility spillover effect in terms of returns is found between gold and USD index. However, there is no transmission of volatility between oil and USD index. Research limitations/Implications: A limitation that could have biased the results is that the returns of gold, oil and USD index were calculated based on the daily prices. In fact, using intraday data could have led to better results in determining the volatility spillover effects between these assets. Moreover, we may incorporate news announcement that may lead to market shocks and could have an impact on the volatility spillover effects under study. Practical implications: Important implications on the risk management practices and portfolio diversification are detected after having the results of the empirical findings. Nowadays the investors are finding it easier to access funds, follow innovative hedging strategies and seek new investment opportunities. However, the result of this dissertation constitutes a perfect proof that there is risk proliferation and volatility transmission in terms of returns among gold, oil and USD index. Yet, knowing this integration and the volatility spillover effect of these assets help investors in portfolio risk reduction. Originality/value: This thesis uses an innovative combination of econometric tools- EGARCH (1, 1) model and 3SLS model, in order to examine volatility spillover effects. Moreover, the use of this basket of models all together in determining the volatility spillover effects added value to this thesis in addition of daily data used for the past 20 years.