This study proposes a new approach to the estimation of the time series properties of daily volatility in financial markets. The estimation technique is a two stage procedure which initially estimates the volatility of any particular trading day from intraday data. This procedure is implemented over a number of trading days to produce a series of daily volatility estimates. A general volatility framework is also developed and the series of daily volatility estimates can be put into this framework to estimate the time series properties of daily volatility. Furthermore, with this new approach it is shown that the time series properties of daily volatility can be modelled in a wide range of functional forms, including those functional forms which capture asymmetric information effects.