some new estimators of integrated volatility一些新的集成波动率的估计.pdf

American Open Journal of Statistics , 2011, 1, 74-80 doi:10.4236/ojs.2011.12008 Published Online July 2011 (http://www.SciRP.org/journal/ojs) Some New Estimators of Integrated Volatility Jaya P. N. Bishwal Department of Mathematics and Statistics , University of North Carolina at Charlotte, Charlotte, USA E-mail : J.Bishwal@ Received April 29, 2011; revised May 28, 2011; accepted June 13, 2011 Abstract We develop higher order accurate estimators of integrated volatility in a stochastic volatility models by using kernel smoothing method and using different weights to kernels. The weights have some relationship to moment problem. As the bandwidth of the kernel vanishes, an estimator of the instantaneous stochastic vola- tility is obtained. We also develop some new estimators based on smoothing splines. Keywords: Stochastic Volatility, Kernel Estimator, Realized Volatility, Moment Problem, Rate of Convergence, Higher Order Asymptotics, Smoothing Spline 1. Introduction This model has been studied in [1]. The integrated volatility is defined as These days high frequency intradaily data of asset re- T 2 V :=  dt. (2.3) turns are available. Hence realized volatility which is a T 0 t measure of the integrated volatility has received consid- In stochasti