Testing the Order of Multivariate Normal Mixture Models

Testing the number of components in multivariate normal mixture models is a long-standing challenge. This paper develops a likelihood-based test of the null hypothesis of M 0 components against the alternative hypothesis of M 0 + 1 components. We derive a local quadratic approximation of the likelihood ratio statistic in terms of the polynomials of the parameters. Based on this quadratic approximation, we propose an EM test of the null hypothesis of M 0 components against the alternative hypothesis of M 0 + 1 components, and derive the asymptotic distribution of the proposed test statistic. The simulations show that the proposed test has good finite sample size and power properties.

• Publication year

  2019

• Venue

  CIRJE F-Series

• Publication date

  2019-02-07

• Fields of study

  Mathematics, Economics

• Identifiers
  arXiv 1902.02920
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Testing the Number of Regimes in Markov Regime Switching Models

  2018cited by this paper
• Testing the Number of Components in Normal Mixture Regression Models

  2015cited by this paper
• A note on the article 'Inference for multivariate normal mixtures' by J. Chen and X. Tan

  2014cited by this paper
• Inference on the Order of a Normal Mixture

  2012influential reference
• Testing the Order of a Finite Mixture

  2010influential reference
• Hypothesis test for normal mixture models: The EM approach

  2009influential reference
• Non-finite Fisher information and homogeneity: an EM approach

  2009cited by this paper
• The likelihood ratio test for general mixture models with or without structural parameter

  2009cited by this paper
• Inference for multivariate normal mixtures

  2008influential reference
• Likelihood‐Based Inference

  2006cited by this paper
• Cluster analysis using multivariate normal mixture models to detect differential gene expression with microarray data

  2006influential reference
• Finite Mixture and Markov Switching Models

  2006cited by this paper
• Asymptotic theory of the likelihood ratio test for the identification of a mixture

  2005cited by this paper
• Hypothesis testing in mixture regression models

  2004influential reference
• Testing for a finite mixture model with two components

  2004cited by this paper
• Asymptotics for likelihood ratio tests under loss of identifiability

  2003influential reference
• Generalized Method of Moments Estimation When a Parameter Is on a Boundary

  2002cited by this paper
• Model-based cluster analysis of microarray gene-expression data

  2002cited by this paper
• The likelihood ratio test for homogeneity in finite mixture models

  2001cited by this paper
• Likelihood ratio test for univariate Gaussian mixture

  2001cited by this paper
• TESTS FOR HOMOGENEITY IN NORMAL MIXTURES IN THE PRESENCE OF A STRUCTURAL PARAMETER

  2000cited by this paper
• Likelihood-based inference with singular information matrix

  2000cited by this paper
• Finite Mixture Models

  2000cited by this paper
• Estimation When a Parameter is on a Boundary

  1999influential reference
• Testing the order of a model using locally conic parametrization : population mixtures and stationary ARMA processes

  1999influential reference
• Cowles Foundation for Research in Economics at Yale University

  1998cited by this paper
• Likelihood ratio tests for genetic linkage

  1997cited by this paper
• Asymptotic distribution of the likelihood ratio test that a mixture of two binomials is a single binomial

  1995cited by this paper
• Mixture models : theory, geometry, and applications

  1995cited by this paper
• Optimal Rate of Convergence for Finite Mixture Models

  1995influential reference
• Empirical Process Methods in Econometrics

  1993influential reference
• Empirical Processes: Theory and Applications

  1990influential reference
• Large sample estimation and hypothesis testing

  1986influential reference
• On the asymptotic performance of the log likelihood ratio statistic for the mixture model and related results

  1984cited by this paper
• On the Use of Discriminant Functions in Taxonomy

  1962influential reference

• Local Asymptotic Distribution Theory for Regime Switching Test in Bilinear Models

  2024cites this paper
• Identification and Estimation of Production Function with Unobserved Heterogeneity

  2023cites this paper
• Testing the Number of Components in Finite Mixture Normal Regression Model with Panel Data

  2022influential citation
• On dynamic choice models, agent's beliefs and unobserved heterogeneity

  2021cites this paper