Articolo in rivista, 2010, ENG, 10.1016/j.sigpro.2009.07.003

Modelling with mixture of symmetric stable distributions using Gibbs sampling

Salas-Gonzalez D.; Kuruoglu E. E.; Ruiz D. P.

Department of Signal Processing, Networking and Communications, University of Granada, Granada, Spain; CNR-ISTI, Pisa, Italy; Department of Applied Physics, University of Granada, Granada, Spain

The stable distribution is a very useful tool to model impulsive data. In this work, a fully Bayesian mixture of symmetric stable distribution model is presented. Despite the non-existence of closed form for alpha-stable distributions, the use of the product property makes it possible to infer on parameters using a straight forward Gibbs sampling. This model is compared to the mixture of Gaussians model. Our proposed methodology is proved to be more robust to outliers than the mixture of Gaussians. Therefore, it is suitable to model mixture of impulsive data. Moreover, as Gaussian is a particular case of the alpha-stable distribution, the proposed model is a generalization of mixture of Gaussians. Mixture of symmetric alpha-stable is intensively tested on both simulated and real data.

Signal processing (Print) 90 (3), pp. 774–783

Keywords

Mixiture distributions, MCMC, Bayesian inference, Alpha-stable distributions, Mixture of stable distributions, Markov chain Monte Carlo, Gibbs sampling

CNR authors

Kuruoglu Ercan Engin

CNR institutes

ISTI – Istituto di scienza e tecnologie dell'informazione "Alessandro Faedo"

ID: 44370

Year: 2010

Type: Articolo in rivista

Last update: 2018-02-06 12:46:22.000

External links

OAI-PMH: Dublin Core

OAI-PMH: Mods

OAI-PMH: RDF

DOI: 10.1016/j.sigpro.2009.07.003

External IDs

CNR OAI-PMH: oai:it.cnr:prodotti:44370

DOI: 10.1016/j.sigpro.2009.07.003

ISI Web of Science (WOS): 000272433400004

Scopus: 2-s2.0-70449519399