@ | ||
@ | ||
cOBiγBεwεw@w€@j | ||
@ | ||
On independent estimation of the diffusion component | ||
@ | ||
We consider independent estimation of the diffusion component of a stochastic process with jumps, leaving other components (drift and pure-jump parts) unknown. The estimation is based on a kind of asymptotically high-frequency discrete data. We propose a very simple estimator of the "long-term realized multipower variation" type. Under appropriate conditions, the estimator is shown to be asymptotically normal with the optimal convergence rate. | ||