FINITE MIXTURES OF MULTIVARIATE SKEW LAPLACE DISTRIBUTIONS

* This paper proposes finite mixtures of multivariate skew Laplace distributions in order to model both skewness and heavy-tailedness in heterogeneous data sets. Maximum likelihood estimators for the parameters of interest are obtained using the EM algorithm. The paper offers a small simulation stud...

Full description

Saved in:
Bibliographic Details
Published inRevstat Vol. 19; no. 1; p. 35
Main Authors Dogru, Fatma Zehra, Bulut, Y. Murat, Arslan, Gloay
Format Journal Article
LanguageEnglish
Published Instituto Nacional de Estatistica 01.01.2021
Instituto Nacional de Estatística | Statistics Portugal
Subjects
Algorithms
Analysis
Data mining
EM algorithm
ML estimation
MSL
multivariate mixture model
Online AccessGet full text

Cover

More Information
Summary:* This paper proposes finite mixtures of multivariate skew Laplace distributions in order to model both skewness and heavy-tailedness in heterogeneous data sets. Maximum likelihood estimators for the parameters of interest are obtained using the EM algorithm. The paper offers a small simulation study and a real data example to illustrate the performance of the proposed mixture model. Keywords: * EM algorithm; ML estimation; multivariate mixture model; MSL. AMS Subject Classification: * 62H12, 65C60.
ISSN:1645-6726
2183-0371
DOI:10.57805/revstat.v19i1.330