Revistes Publicacions URV: SORT - Statistics and Operations Research Transactions> 2016

Log-ratio methods in mixture models for compositional data sets

  • Datos identificativos

    Identificador: RP:2450
    Handle: https://hdl.handle.net/20.500.11797/RP2450
    Autores:
    Mateu-Figueras, GlòriaMartín-Fernández, Josep AntoniComas-Cufí, Marc
    Resumen:
    When traditional methods are applied to compositional data misleading and incoherent results could be obtained. Finite mixtures of multivariate distributions are becoming increasingly important nowadays. In this paper, traditional strategies to fit a mixture model into compositional data sets are revisited and the major difficulties are detailed. A new proposal using a mixture of distributions defined on orthonormal log-ratio coordinates is introduced. A real data set analysis is presented to illustrate and compare the different methodologies.

  • Otros:

    Autor/es de la URV: Mateu-Figueras, Glòria Martín-Fernández, Josep Antoni Comas-Cufí, Marc
    Palabras clave: Compositional data, Finite Mixture, Log ratio, Model-based clustering, Normal distribution, Orthonormal coordinates, Simplex
    Resumen: When traditional methods are applied to compositional data misleading and incoherent results could be obtained. Finite mixtures of multivariate distributions are becoming increasingly important nowadays. In this paper, traditional strategies to fit a mixture model into compositional data sets are revisited and the major difficulties are detailed. A new proposal using a mixture of distributions defined on orthonormal log-ratio coordinates is introduced. A real data set analysis is presented to illustrate and compare the different methodologies.
    Año de publicación de la revista: 2016
    Tipo de publicación: info:eu-repo/semantics/publishedVersion info:eu-repo/semantics/article

  • Palabras clave:

    Compositional data, Finite Mixture, Log ratio, Model-based clustering, Normal distribution, Orthonormal coordinates, Simplex

  • Documentos:

    DocumentPrincipal

  • Cerca a google

    Search to google scholar