Estimation and model selection for model-based clustering with the conditional classification likelihood
From MaRDI portal
Abstract: The Integrated Completed Likelihood (ICL) criterion has been proposed by Biernacki et al. (2000) in the model-based clustering framework to select a relevant number of classes and has been used by statisticians in various application areas. A theoretical study of this criterion is proposed. A contrast related to the clustering objective is introduced: the conditional classification likelihood. This yields an estimator and a model selection criteria class. The properties of these new procedures are studied and ICL is proved to be an approximation of one of these criteria. We oppose these results to the current leading point of view about ICL, that it would not be consistent. Moreover these results give insights into the class notion underlying ICL and feed a reflection on the class notion in clustering. General results on penalized minimum contrast criteria and on mixture models are derived, which are interesting in their own right.
Recommendations
- Choosing the number of clusters in a finite mixture model using an exact integrated completed likelihood criterion
- Variable selection for model-based clustering using the integrated complete-data likelihood
- Classification and models
- Model selection for Gaussian mixture models
- scientific article; zbMATH DE number 2034563
Cites work
- A non asymptotic penalized criterion for Gaussian mixture model selection
- Asymptotic Statistics
- Choosing starting values for the EM algorithm for getting the highest likelihood in multivariate Gaussian mixture models
- Clustering Criteria and Multivariate Normal Mixtures
- Concentration inequalities and model selection. Ecole d'Eté de Probabilités de Saint-Flour XXXIII -- 2003.
- Consistent estimation of the order of mixture models.
- Estimating the dimension of a model
- Exact posterior distributions and model selection criteria for multiple change-point detection problems
- Finite mixture models
- Gaussian model selection
- scientific article; zbMATH DE number 3567782 (Why is no real title available?)
- scientific article; zbMATH DE number 1219611 (Why is no real title available?)
- scientific article; zbMATH DE number 3444596 (Why is no real title available?)
- Maximum likelihood principle and model selection when the true model is unspecified
- Methods for merging Gaussian mixture components
- Minimal penalties for Gaussian model selection
- Mixture Densities, Maximum Likelihood and the EM Algorithm
- Model selection by resampling penalization
- Model-based clustering and classification with non-normal mixture distributions
- Model-Based Clustering, Discriminant Analysis, and Density Estimation
- Nonparametric estimation of regression level sets using kernel plug-in estimator
- Numerical Analysis for Statisticians
- Risk bounds for model selection via penalization
- Selecting models focussing on the modeller's purpose
- Slope heuristics: overview and implementation
- Statistical analysis of finite mixture distributions
- Uncovering latent structure in valued graphs: a variational approach
- Uniform Central Limit Theorems
Cited in
(19)- Modeling heterogeneous peer assortment effects using finite mixture exponential random graph models
- Probability of misclassification in model-based clustering
- Model-based clustering
- Finding the Number of Normal Groups in Model-Based Clustering via Constrained Likelihoods
- Choosing the number of clusters in a finite mixture model using an exact integrated completed likelihood criterion
- Poisson Kernel-Based Clustering on the Sphere: Convergence Properties, Identifiability, and a Method of Sampling
- Identifying the number of components in Gaussian mixture models using numerical algebraic geometry
- Penalized model-based clustering of complex functional data
- Selecting the number of clusters, clustering models, and algorithms. A unifying approach based on the quadratic discriminant score
- Order selection with confidence for finite mixture models
- Functional Mixed Effects Clustering with Application to Longitudinal Urologic Chronic Pelvic Pain Syndrome Symptom Data
- A survey on model-based co-clustering: high dimension and estimation challenges
- Model-based clustering with missing not at random data
- Mixture of longitudinal factor analyzers and their application to the assessment of chronic pain
- Selection of number of clusters and warping penalty in clustering functional electrocardiogram
- Functional zoning of biodiversity profiles
- PanIC: consistent information criteria for general model selection problems
- Model selection for Gaussian latent block clustering with the integrated classification likelihood
- Bayesian nonparametric clustering as a community detection problem
This page was built for publication: Estimation and model selection for model-based clustering with the conditional classification likelihood
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2346523)
