Stylized Facts and Dynamic Modeling of High-frequency Data on Precious Metals

Taking advantage of a trades-and-quotes database, the main stylized facts and dynamic properties of a time series related to spot precious metals, that is, gold, silver, palladium, and platinum, are documented. The behavior of spot prices, returns, volume, and selected liquidity measures is analyzed. A clear evidence of periodic patterns matching the trading hours of the most active markets, London, Zurich, New York, as well as Asian markets, is found. The time series of spot returns have thus properties similar to those of traditional financial assets with fat tails, asymmetry, periodic behaviors in the conditional variances, and volatility clustering. The empirical analyzes show, as expected, that gold is the most liquid and less volatile asset, whereas palladium and platinum are traded less.