Volatility Estimation and Jump Testing via Realized Information Variation

We put forward two jump-robust estimators of integrated volatility, namely realized information variation (RIV) and realized information power variation (RIPV). The "information" here refers to the difference between two-grid of ranges in high-frequency intervals, which preserves continuous variation and eliminates jump variation asymptotically. We give several probabilistic laws to show that RIV is much more efficient than most of the other estimators, e.g. 8.87 times more efficient than bi-power variation, and RIPV has a fast jump convergence rate at Op(1/n), while the others are usually Op(1/sqrt(n)) in the literature. We also extend our results to integrated quarticity and higher-order variation estimation, and then propose a new jump testing method. Simulation studies provide extensive evidence on the finite sample properties of our estimators and tests, comparing with alternative methods. The simulations support our theoretical results on volatility estimation and demonstrate that our jump testing method has much lower type I error for smaller sample frequencies, or in the presence of microstructure noise.