Spot estimation for fractional Ornstein-Uhlenbeck stochastic volatility model: consistency and central limit theorem (Q2194053)

The article refers to a rough stochastic volatility model with fractional Ornstein-Uhlenbeck log-volatility and Itô-semimartingale log-price processes. The important contribution of the paper is the derivation, under some very special assumptions, of a central limit theorem (CLT) with a better convergence rate -- as in mentioned previous works on this subject -- for estimators of fractional Ornstein-Uhlenbeck volatility driven by a fractional Brownian motion with Hurst parameter \(H\in(0,1)\). The log-price of a financial asset is an Itô-semimartingale observed at a discretization scheme, the volatility process is not observed and is estimated, the log-volatility follows a fractional Ornstein-Uhlenbeck process and for the estimation non-truncated and truncated estimators are used. After a presentation of introductory elements and assumptions in the second section, the main central limit theorem and corollaries on asymptotic confidence interval for volatility estimators are shown in the third section and the proofs developed in the fourth section. The last section is dedicated to simulation results.