Time-varying mixture MEM for realized volatility

TL;DRAbstract

In this paper, we model the dynamics of realized volatility as a multiplicative error model with a mixture of distributions for the innovation term. The mixture is usually justified to capture the right tail behavior of the innovation more accurately, thus providing a potentially better fit in all the cases where the density forecast of the realized volatility is needed and considerations about its variability are appropriate. The model provides a time-varying volatility of volatility behavior. An application is provided for the Johnson’s and Johnson’s stock for the period 2001-2006.

Chat with Paper

AI Agents for this Paper