Econometric analysis of realised volatility and its use in estimating stochastic volatility models

The availability of intra-day data on the prices of speculative assets means that we can use quadratic variation like measures of activity in financial markets, called realised volatility, to study the stochastic properties of returns. Here we derive the moments and the asymptotic distribution of the realised volatility error - the difference between realised volatility and the actual volatility. These properties can be used to allow us to estimate the parameters of stochastic volatility models.