Description:
This paper proposes an extended multivariate EGARCH model for multivariate volatility modeling that addresses several limitations of existing models. Specifically, it overcomes the zero-return problem and allows for negative news and volatility spillover effects, making it a promising tool for modeling multivariate volatility. While the QML estimator has some limitations, such as non-invertibility and unclear asymptotic properties, Monte Carlo simulations suggest that it is consistent and asymptotically normal for larger sample sizes (i.e., T ≥ 2500). An empirical example demonstrates the model’s superior performance compared to multivariate GARCH and Log-GARCH models in investigating volatility spillover effects among the bond, stock, crude oil, and gold markets