Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
6870135 | Computational Statistics & Data Analysis | 2014 | 15 Pages |
Abstract
The Fourier estimator of Malliavin and Mancino depends on both sample size and a so-called cutting frequency. The latter controls the number of Fourier coefficients to be included, and it also determines how the Fourier estimator responds to market microstructure noise. By examining the finite sample properties of the Fourier estimator, an easy-to-implement procedure is developed for the optimal cutting frequency which minimizes the mean squared error in the presence of the microstructure noise, along with a modified Whittle likelihood approach for the estimation of the signal-to-noise ratio.
Keywords
Related Topics
Physical Sciences and Engineering
Computer Science
Computational Theory and Mathematics
Authors
Fangfang Wang,