Share Email Print
cover

Proceedings Paper

Class of time-domain procedures for testing that a stationary time series is Gaussian
Author(s): Eric Moulines; Karim Choukri; Jean-Francois Cardoso
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

In this contribution, a class of time-domain procedures for testing that a stationary time-series is Gaussian, is presented. These tests are based on minimum chi-square statistics in the deviations of certain sample statistics from their ensemble counterpart. Exact asymptotic distributions of these tests are derived under the null hypothesis of Gaussianity and under a class of local and fixed alternatives. Two specific tests are then developed, based respectively on the third-order and the fourth-order moments and on the characteristic functions. Extensive simulations are presented to illustrate the power of the test against various alternatives (including additive and non-additive contaminations and non-linear serial dependence.

Paper Details

Date Published: 28 October 1994
PDF: 12 pages
Proc. SPIE 2296, Advanced Signal Processing: Algorithms, Architectures, and Implementations V, (28 October 1994); doi: 10.1117/12.190829
Show Author Affiliations
Eric Moulines, Telecom Paris (France)
Karim Choukri, Telecom Paris (France)
Jean-Francois Cardoso, Telecom Paris (France)


Published in SPIE Proceedings Vol. 2296:
Advanced Signal Processing: Algorithms, Architectures, and Implementations V
Franklin T. Luk, Editor(s)

© SPIE. Terms of Use
Back to Top