Volatility dynamics and heterogeneous markets

Recent research has suggested that intra‐day volatility may possess a component structure, resulting either from the arrival of heterogeneous information or the actions of heterogeneous market agents. This paper reports direct evidence for the existence of such components in S&P500 index and DM/$ exchange rate data. Estimation of a FIGARCH model supports the contention that volatility dynamics result from multiple sources. Using a HARCH conditional variance model which defines volatility components over differing time horizons, confirmatory evidence of heterogeneous components is reported, in which context the impact of high‐frequency speculation and noise‐trading are particularly apparent. Copyright © 2006 John Wiley & Sons, Ltd.