Conditional Heavy Tails, Volatility Clustering and Asset Prices of the Precious Metal

The world palladium demand has increased steadily and dramatically in the years of 21th century. However, its spot price still has not reached the peak level observed in January 17, 2001. In 2008, a single-day increase of the palladium spot price has exceeded 37%, which witnesses significant risk for investments in the world palladium market. In this paper, we apply the GARCH model with heavy-tailed distributions into the palladium spot returns series for risk management purpose. We compare empirical performance of the Student’s t distribution and the normal reciprocal inverse Gaussian (NRIG) distribution. Our results show the newly-developed distribution, the NRIG, cannot outperform the older fashion one, the Student’s t distribution. Nevertheless, our results do demonstrate that it is important to incorporate conditional heavy tails for precious metals’ spot return modelling.