Data-driven penalty calibration: A case study for Gaussian mixture model selection

DOI10.1051/ps/2010002zbMath1395.62163OpenAlexW2089460110MaRDI QIDQ4918495

Bertrand Michel, Cathy Maugis

Publication date: 25 April 2013

Published in: ESAIM: Probability and Statistics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1051/ps/2010002

zbMATH Keywords

model-based clusteringslope heuristicspenalized likelihood criterionnoisy variable detection

Mathematics Subject Classification ID

Density estimation (62G07) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Related Items (7)

Dealing with overdispersion in multivariate count data ⋮ Clustering and variable selection for categorical multivariate data ⋮ On the nonparametric maximum likelihood estimator for Gaussian location mixture densities with application to Gaussian denoising ⋮ Nonparametric finite translation hidden Markov models and extensions ⋮ Comments on: \(\ell_{1}\)-penalization for mixture regression models ⋮ A non-asymptotic approach for model selection via penalization in high-dimensional mixture of experts models ⋮ Stable and visualizable Gaussian parsimonious clustering models

Uses Software

• Mixmod

This page was built for publication: Data-driven penalty calibration: A case study for Gaussian mixture model selection