Novel robust gaussianity test for sparse data

Document Type

Conference Proceeding

Publication Date

1-1-2010

Abstract

In this paper, a fundamental but important statistical signal processing characteristic, namely the Gaussianity or normality, is studied. In contrast to the existing conventional Gaussianity measures, we propose a novelmeasure, which is based on Kullback-Leibler divergence (KLD) between the Gaussian probability density function (PDF) and the generalized Gaussian PDF incorporated with the skewness for the normality test. In our studies, conventional normality tests may often not be robust when they are employed for the non-Gaussian processes with symmetric PDFs. We call this new test as the KGGS test. Our proposed KGGS test is heuristically justi-fied to be more robust than the conventional tests for different PDFs, especially for the symmetric PDFs. ©2010 IEEE.

Publication Source (Journal or Book title)

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

First Page

3914

Last Page

3917

This document is currently not available here.

Share

COinS