Combining multiple frequencies in Realized GARCH models

This paper proposes extensions of the Realized GARCH model of Hansen et al. (2012) by incorporating information from multiple realized volatility measures computed at different sampling frequencies to achieve an optimal trade-off between bias and efficiency. Future volatility forecasts are determined by a weighted average of the considered realized measures, where the weights are time-varying and adaptively determined according to the estimated amount of noise and jumps. This specification aims to reduce bias effects related to the different sampling frequencies at which the realized measure are computed.