2026-05-05T20:43:31Zhttps://uvadoc.uva.es/oai/request

oai:uvadoc.uva.es:10324/214122021-06-23T10:09:46Zcom_10324_1151com_10324_931com_10324_894col_10324_1278

García Escudero, Luis Ángel Gordaliza Ramos, Alfonso Greselin, Francesca Ingrassia, Salvatore Mayo Iscar, Agustín 2016-12-01T23:01:50Z 2016-12-01T23:01:50Z 2016 Computational Statistics and Data Analysis, vol. 99, pp. 131-147. ISSN: 0167-9473 http://uvadoc.uva.es/handle/10324/21412 10.1016/j.csda.2016.01.005 131 147 Computational Statistics and Data Analysis, 99 Producción Científica 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. 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. application/pdf eng ELSEVIER info:eu-repo/semantics/restrictedAccess Springer Constrained estimation, Factor analyzers modeling, Mixture models, Model-based clustering, Robust estimation The joint role of trimming and constraints in robust estimation for mixtures of Gaussian factor analyzers. info:eu-repo/semantics/article http://www.sciencedirect.com/science/article/pii/S0167947316000141 SI