Analyzing and Exploiting Asymmetries in the News Impact Curve

The recently proposed class of MixN-GARCH models, which couple a mixed normal distributional structure with linked GARCH-type dynamics, has been shown to offer a plausible decomposition of the contributions to volatility, as well as admirable out-of-sample forecasting performance, for financial asset returns. A general feature of these processes are constant mixing weights of the component densities, which leads to tractable stationarity conditions, but may not be a realistic assumption in general. This paper relaxes this assumption by considering different specifications of time-varying mixing weights for the MixN-GARCH model class. In particular, by relating current weights to past returns via sigmoid response functions, an empirically more realistic representation of Engle and Ng's (1993) news impact curve with an asymmetric impact of unexpected return shocks on future volatility is obtained, and large gains in terms of in-sample fit and out-of-sample VaR forecasting performance can be realized.