Volatility asymmetry in high frequency data

The paper examines the important stylized fact of asymmetric volatility in high frequency data. It demonstrates that one of the two main factors explaining this phenomenon, the leverage effect, is measured more precisely using higher frequency data. The paper also examines the lead-lag relation between volatility and returns in high frequency data. The impact of returns shocks on future squared returns is shown to decay geometrically and remain significant for at least 3 days. A temporal aggregation formula for the correlation between squared returns and lagged returns that permits more accurate measures of the daily correlation using high frequency data is also derived. Lastly, the paper examines the ability of one and two factor stochastic volatility models to explain the pattern observed in high frequency data.