Ítem


An EM-Algorithm Based Method to Deal with Rounded Zeros in Compositional Data under Dirichlet Models

Zeros in compositional data are classified into “rounded” zeros and “essential” zeros. Therounded zero corresponds to a small proportion or below detection limit value while the essentialzero is an indication of the complete absence of the component in the composition. Several parametricand non-parametric imputation techniques have been proposed to replace rounded zeros andmodel the essential zeros under logratio model. In this paper, a new method based on EM algorithmis proposed for replacing rounded zeros. The proposed method is illustrated using simulateddata

Universitat de Girona. Departament d’Informàtica i Matemàtica Aplicada

Altres contribucions: Universitat de Girona. Departament d’Informàtica i Matemàtica Aplicada
Autor: Hijazi, R.
Resum: Zeros in compositional data are classified into “rounded” zeros and “essential” zeros. Therounded zero corresponds to a small proportion or below detection limit value while the essentialzero is an indication of the complete absence of the component in the composition. Several parametricand non-parametric imputation techniques have been proposed to replace rounded zeros andmodel the essential zeros under logratio model. In this paper, a new method based on EM algorithmis proposed for replacing rounded zeros. The proposed method is illustrated using simulateddata
Accés al document: http://hdl.handle.net/2072/273635
Llenguatge: eng
Editor: Universitat de Girona. Departament d’Informàtica i Matemàtica Aplicada
Drets: Tots els drets reservats
Matèria: Estadística matemàtica -- Congressos
Mathematical statistics -- Congresses
Anàlisi multivariable -- Congressos
Mathematical statistics -- Congresses
Títol: An EM-Algorithm Based Method to Deal with Rounded Zeros in Compositional Data under Dirichlet Models
Tipus: info:eu-repo/semantics/conferenceObject
Repositori: Recercat

Matèries

Autors