Uniform and Lp Convergences for Nonparametric Kernel Estimation of Diffusion Models
报告人： Bin Wang, Harbin Institute of Technology, Shenzhen
时间：2017-10-26 14:00 ~ 15:00
地点：217, Guanghua Building 2
Abstract:
We establish the uniform convergence rates of, nonparametric kernel estimators of the local time, kernel estimators and their derivatives of the drift and volatility functions, for a discretely sampled diffusion process that is possibly nonstationary. Our results are derived using two-dimensional asymptotics with time span increasing to infinity and sampling interval shrinking to zero simultaneously. We also propose modified kernel estimators of drift and volatility functions, which are smooth versions of truncated kernel estimators, and provide the Lp convergence rates of those modified kernel estimators as well as their derivatives. Our convergence results are useful for non/semiparametric inference for recurrent diffusion processes.
About the Speaker:
Dr. Bin Wang is currently an assistant professor at Harbin Institute of Technology, Shenzhen. He obtained his Ph.D. degree in Economics from Indiana University in 2012. His research interests are Econometric, Time Series, and Financial Econometrics. He has some working papers with second round revisions requested by journals including Journal of Econometrics, Econometric Theory.
