Exponential GARCH Modeling With Realized Measures of Volatility

We introduce the realized exponential GARCH model that can use multiple realized volatility measures for the modeling of a return series. The model specifies the dynamic properties of both returns and realized measures, and is characterized by a flexible modeling of the dependence between returns and volatility. We apply the model to 27 stocks and an exchange traded fund that tracks the S&P 500 index and find specifications with multiple realized measures that dominate those that rely on a single realized measure. The empirical analysis suggests some convenient simplifications and highlights the advantages of the new specification.Center for Research in Econometric Analysis of Time Series, CREATES; Danish National Research Foundation; National Natural Science Foundation of China [71201001]SCI(E)SSCIARTICLEpeter.hansen@eui.eu; zhuohuang@nsd.pku.edu.cn2269-2873