Testing Nonstationary Time Series for Gaussianity and Linearity using the Evolutionary Bispectrum: An application to Internet Traffic Data

Subba Rao, T and Tsolaki, E (2006) Testing Nonstationary Time Series for Gaussianity and Linearity using the Evolutionary Bispectrum: An application to Internet Traffic Data. [MIMS Preprint]

PDF psrr07-2005.pdf Download (459kB)

Abstract

We propose statistical tests for Gaussianity and linearity of nonstationary time series based on the evolutionary bispectrum. These tests can be applied to a particular subclass of nonstationary processes, the so-called oscillatory (also known as slowly-varying) processes. We then apply these tests to time series of network measurements arising from internet traffic. Recent works by several researchers have demonstrated that such internet trafic processes are typically nonstationary. Also, the question of whether such processes can be described by some model whose parameters vary with time has been raised and studied at some length. We use the tests developed in this paper to show that there is evidence of non Gaussianity and nonlinearity in such processes uner the assumption that they are described by a model whose parameters (and so its spectral characteristics) vary slowly with time.

Item Type:	MIMS Preprint
Subjects:	MSC 2010, the AMS's Mathematics Subject Classification > 62 Statistics
Depositing User:	Dr Peter Neal
Date Deposited:	24 Mar 2006
Last Modified:	08 Nov 2017 18:18
URI:	https://eprints.maths.manchester.ac.uk/id/eprint/192

Actions (login required)

View Item