Heterogeneous Asymmetric Dynamic Conditional Correlation Model with Stock Return and Range

Abstract

The paper considers the return and range model with dynamic conditional correlations (DCC). The paper suggests the new speci cations for the asymmetric effects on log-volatilities and dynamic correlations, combined with long-run dependences. The new DCC model can be estimated by the quasi-maximum likelihood method. The empirical analysis on Nikkei 225, Hang Seng and Straits Times indices shows the daily, weekly and monthly pattern of the asymmetric effects. For the period including the global nancial crisis, the new DCC model provides plausible one-step-ahead forecasts of the VaR thresholds, and yields positive economic values of switching from other DCC models.