Statistics and Its Interface

Volume 1 (2008)

Number 2

Spot volatility estimation for high-frequency data

Pages: 279 – 288

DOI: https://dx.doi.org/10.4310/SII.2008.v1.n2.a5

Authors

Jianqing Fan (Princeton University, Princeton, New Jersey, U.S.A.)

Yazhen Wang (University of Connecticut, Storrs, Conn., U.S.A.)

Abstract

The availability of high-frequency intraday data allows us to accurately estimate stock volatility. This paper employs a bivariate diffusion to model the price and volatility of an asset and investigates kernel type estimators of spot volatility based on high-frequency return data. We establish both pointwise and global asymptotic distributions for the estimators.

Keywords

asymptotic normality, CIR model, constant elasticity of diffusion, extreme distribution, kernel estimator, long memory, stock price

Published 1 January 2008