Surprising information, the MDH, and the relationship between volatility and trading volume

This paper explains the concept of surprising information with a sign effect. Employing the mixture of distribution hypothesis (MDH), this paper also theoretically demonstrates that the effect of surprising information on the relationship between volatility and trading volume contrasts with that of general information, and proposes a method to detect the unobservable surprising information. Furthermore, incorporating surprising information with a sign effect, this paper suggests an information-type switching GARCH-V model. Strong evidence in favor of the model specification over the standard GARCH models is based on empirical application with high frequency data, supporting the dependence of the relationship between volatility and trading volume on the type of information. (C) 2010 Elsevier B.V. All rights reserved.