Skip navigation
Por favor, use este identificador para citar o enlazar este ítem: http://uvadoc.uva.es/handle/10324/21412
Título: The joint role of trimming and constraints in robust estimation for mixtures of Gaussian factor analyzers.
Autor: García-Escudero, Luis Ángel
Gordaliza, Alfonso
Greselin, Francesca
Ingrassia, Salvatore
Mayo Iscar, Agustín
Año del Documento: 2016
Editorial: ELSEVIER
Descripción: Producción Científica
Documento Fuente: Computational Statistics and Data Analysis, vol. 99, pp. 131-147.
Resumen: Mixtures of Gaussian factors are powerful tools for modeling an unobserved heterogeneous population, offering – at the same time – dimension reduction and model-based clustering. The high prevalence of spurious solutions and the disturbing effects of outlying observations in maximum likelihood estimation may cause biased or misleading inferences. Restrictions for the component covariances are considered in order to avoid spurious solutions, and trimming is also adopted, to provide robustness against violations of normality assumptions of the underlying latent factors. A detailed AECM algorithm for this new approach is presented. Simulation results and an application to the AIS dataset show the aim and effectiveness of the proposed methodology.
Materias (normalizadas): Constrained estimation, Factor analyzers modeling, Mixture models, Model-based clustering, Robust estimation
ISSN: ISSN: 0167-9473
Revisión por Pares: SI
DOI: 10.1016/j.csda.2016.01.005
Patrocinador: Ministerio de Economía y Competitividad and FEDER, grant MTM2014-56235-C2-1-P, and by Consejería de Educación de la Junta de Castilla y León, grant VA212U13, by grant FAR 2015 from the University of Milano-Bicocca and by grant FIR 2014 from the University of Catania.
Version del Editor: http://www.sciencedirect.com/science/article/pii/S0167947316000141
Propietario de los Derechos: Springer
Idioma: eng
URI: http://uvadoc.uva.es/handle/10324/21412
Derechos: info:eu-repo/semantics/restrictedAccess
Aparece en las colecciones:DEP24 - Artículos de revista

Ficheros en este ítem:
Fichero Descripción TamañoFormato 
GGGIM_reviewedAlan.pdf921,03 kBAdobe PDFThumbnail
Visualizar/Abrir


Los ítems de UVaDOC están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.

Comentarios
Universidad de Valladolid
Powered by MIT's. DSpace software, Version 5.5