Fourier Analysis of Serial Dependence Measures

15 Pages Posted: 14 Dec 2017

See all articles by Ria Van Hecke

Ria Van Hecke

Ruhr University of Bochum

Stanislav Volgushev

Ruhr University of Bochum - Faculty of Mathematics

Holger Dette

Ruhr University of Bochum - Faculty of Mathematics

Date Written: January 2018

Abstract

Classical spectral analysis is based on the discrete Fourier transform of the autocovariances. In this article we investigate the asymptotic properties of new frequency‐domain methods where the autocovariances in the spectral density are replaced by alternative dependence measures that can be estimated by ‐ statistics. An interesting example is given by Kendall's, for which the limiting variance exhibits a surprising behavior.

Keywords: Spectral theory, strictly stationary time series, ‐statistics

Suggested Citation

Van Hecke, Ria and Volgushev, Stanislav and Dette, Holger, Fourier Analysis of Serial Dependence Measures (January 2018). Journal of Time Series Analysis, Vol. 39, Issue 1, pp. 75-89, 2018, Available at SSRN: https://ssrn.com/abstract=3087660 or http://dx.doi.org/10.1111/jtsa.12266

Ria Van Hecke (Contact Author)

Ruhr University of Bochum ( email )

Faculty of Management and Economics
Chair of International Economics
Bochum, 44780
Germany

Stanislav Volgushev

Ruhr University of Bochum - Faculty of Mathematics

D-44780 Bochum
Germany

Holger Dette

Ruhr University of Bochum - Faculty of Mathematics ( email )

Universitatsstr. 150
D-44780 Bochum
Germany
+ 49 234 700 3284 (Phone)
+ 49 234 7094 559 (Fax)

Do you want regular updates from SSRN on Twitter?

Paper statistics

Downloads
0
Abstract Views
142
PlumX Metrics