An exact discretization of continuous time stochastic volatility processes observed at irregularly spaced times is used to give insights on how a coherent GARCH model can be specified for such data. The relation of our approach with those in the existing literature is studied

We present a unified and up-to-date overview of temporal aggregation techniques for univariate and multivariate time series models explaining in detail, although intuitively, the technical machinery behind the results. Some empirical applications illustrate the main issues