Statistics Seminar: "Optimal Kernel Estimation of Spot Volatility with Leverage and Microstructure Noise"

Speaker: Bei Wu, Washington University in St. Louis

Abstract: Continuous Ito semi-martingale models for the dynamics of asset returns have been widely used in financial econometric. A key component of the model, the spot volatility, plays a crucial rule in option pricing, portfolio management, and financial risk assessment. In this work, we develop a nonparametric kernel method to estimate the spot volatility for high frequency data in the presence of market microstructure noise and leverage effects.  We prove consistency and feasible central limit theorems (CLT). Compared to existing methods, our estimator achieves optimal convergence rate. As an application of the CLT, we obtain results for the optimal tuning of the estimators in regard to its bandwidth and kernel selections. We proceed to propose an iterative method to implement the optimal bandwidth parameter of the estimator.  In Monte Carlo simulations, we compare the finite sample performance of our estimator with other existing estimators. This is based on joint work with Dr. Figueroa-Lopez.


Host: Jose Figueroa-Lopez