Abstract:
|
The multiplicative factor multi-frequency component (MF)2 GARCH process proposed by Conrad and Engle in 2021, considers conditional variance as the product of a short-term volatility component modeled as a GJR-GARCH process and a long-term component expressed by a multiplicative error model (MEM) for the past forecast errors of the GARCH component. In this paper, we propose an exponential GARCH version of the (MF)2 formulation, which models the short-term volatility component as a less restrictive EGARCH process and the long-term component as a MEM. Further, the long-term model is formulated to account for a differential effect of the long-term accumulation of positive and negative shocks. Small sample properties of the parameter estimates are studied using a Monte-Carlo simulation and the utility of the proposed model is illustrated using a real-life data set.
|