Expectation–Maximization algorithm for finite mixture of α-stable distributions

A Gaussian Mixture Model (GMM) is a parametric probability density function built as a weighted sum of Gaussian distributions. Gaussian mixtures are used for modelling the probability distribution in many fields of research nowadays. Nevertheless, in many real applications, the components are skewed...

Full description

Saved in:

Bibliographic Details
Published in:	Neurocomputing (Amsterdam) Vol. 413; pp. 210 - 216
Main Authors:	Castillo-Barnes, D., Martinez-Murcia, F.J., Ramírez, J., Górriz, J.M., Salas-Gonzalez, D.
Format:	Journal Article
Language:	English
Published:	Elsevier B.V 06.11.2020
Subjects:	Alpha-stable distribution Alpha-stable mixture model Expectation–Maximization algorithm Expectation–Maximization algorithm Alpha-stable mixture model Alpha-stable distribution
ISSN:	0925-2312, 1872-8286
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	A Gaussian Mixture Model (GMM) is a parametric probability density function built as a weighted sum of Gaussian distributions. Gaussian mixtures are used for modelling the probability distribution in many fields of research nowadays. Nevertheless, in many real applications, the components are skewed or heavy tailed. For that reason, it is useful to model the mixtures as components with α-stable distribution. In this work, we present a mixture of skewed α-stable model where the parameters are estimated using the Expectation–Maximization algorithm. As the Gaussian distribution is a particular limiting case of α-stable distribution, the proposed model is a generalization of the widely used GMM. The proposed algorithm is much faster than the parameter estimation of the α-stable mixture model using a Bayesian approach and Markov chain Monte Carlo methods. Therefore, it is more suitable to be used for large vector observations.
ISSN:	0925-2312 1872-8286
DOI:	10.1016/j.neucom.2020.06.114