A predictive deviance criterion for selecting a generative model in semi-supervised classification

DOI10.1016/j.csda.2013.02.010zbMath1468.62202OpenAlexW2011795890MaRDI QIDQ1800092

Gilles Celeux, Gérard Govaert, Christophe Biernacki, Vincent Vandewalle

Publication date: 19 October 2018

Published in: Computational Statistics and Data Analysis (Search for Journal in Brave)

Full work available at URL: https://hal.inria.fr/inria-00516991/file/RR-7377.pdf

zbMATH Keywords

EM algorithmmaximum likelihoodAICBICGaussian mixture modelsBECinformation criteriagenerative modelscross-validated error rate

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Related Items

The importance of being clustered: uncluttering the trends of statistics from 1970 to 2015

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Estimating the dimension of a model
• Asymptotically minimax regret procedures in regression model selection and the magnitude of the dimension penalty.
• Finite mixture and Markov switching models.
• A Generalized Discriminant Rule When Training Population and Test Population Differ on Their Descriptive Parameters
• The General Distribution of the Error Rate of a Classification Procedure with Application to Logistic Regression Discrimination
• Consequences and Detection of Misspecified Nonlinear Regression Models
• Logistic Discrimination and Bias Correction in Maximum Likelihood Estimation
• Asymptotic Statistics
• Regularized Gaussian Discriminant Analysis Through Eigenvalue Decomposition
• Model-Based Clustering, Discriminant Analysis, and Density Estimation
• Regression Analysis when the Dependent Variable Is Truncated Normal
• Asymptotic Properties of Non-Linear Least Squares Estimators