In-sample Asymptotics and Across-sample Efficiency Gains for High Frequency Data Statistics
Joint with: Eric Ghysels and Per Mykland
We revisit the widely used in-sample asymptotic analysis developed by Jacod (1994) and Barndorf-Nielsen and Shephard (2002) and extensively used in the realized volatility literature. We show that there are gains to be made in estimating current realized volatility from considering realizations in prior periods. Our analysis is reminiscent of local-to-unity asymptotics of Bobkoski (1983), Phillips (1987), Chan and Wei (1987) among many others.
For more information, contact Frank Schorfheide.