CMStatistics 2016: Start Registration
View Submission - CFE
Title: Stationary bootstrapping for realized covolatility with high frequency noisy and asynchronous observations. Authors:  Dong Wan Shin - Ewha university (Korea, South)
Eunju Hwang - Gachon University (Korea, South) [presenting]
Abstract: Under two important modern financial market features such as noise and non-synchronicity, the average realized volatility matrix is adopted for consistent estimators of the integrated covariations, and their asymptotic normal theories are established. Furthermore, stationary bootstrapping is applied to propose the bootstrap realized volatility matrix and its validity for the high frequency heterogeneous returns is proved by showing that there exist parameters of the stationary bootstrap blocks so that the bootstrap consistencies hold. Confidence intervals and hypothesis tests for the integrated covariance, regression coefficient and correlation coefficient are constructed. The proposed bootstrap methods extend the existing i.i.d. bootstrapping methods for realized covariations that are confined to synchronous noise-free sampling. A Monte Carlo experiment shows good finite sample performances of the proposed bootstrap methods for high frequency noisy asynchronous samples.