Results 1 - 1 of 1
Results 1 - 1 of 1. Search took: 0.017 seconds
[en] It is a well-known stylized property that statistical distributions of financial time-series exhibit significant variation over time, as well as prevalent deviations from normality. The vast majority of earlier financial econometric studies has employed GARCH and SV model-classes in order to capture the empirically-observed time-variation (clustering) in volatility. This has also offered a flexible parametric framework to address other empirically-observed properties of the data, such as leverage and/or feedback effects -among others- which are partly responsible for inducing asymmetries in the distributions of returns. Another major source of asymmetry and tail-heaviness has been attributed to the existence of latent jump dynamics in both returns and variance. Recent advances to the econometrics literature of high-frequency data have popularized approaches that permit identification of the underlying jump dynamics. These realized jump-driven estimators carry significant information not only for the returns, but also for volatility.