A note on ‘Testing the number of components in a normal mixture’

Author: Jeffries N.O.

Source: Biometrika, Volume 90, Number 4, December 2003 , pp. 991-994(4)

Buy & download fulltext article:

Price: $35.78 plus tax (Refund Policy)

Abstract:

In a recent paper, Lo et al. (2001) propose a test for the likelihood ratio statistic based on the Kullback–Leibler information criterion when testing the null hypothesis that a random sample is drawn from a mixture of k<inf>0</inf> normal components against the alternative hypothesis of a mixture with k<inf>1</inf> normal components with k<inf>0</inf> less than k<inf>1</inf>. However, this result requires conditions that are generally not met when the null hypothesis holds. Consequently, the result is not proven and simulations suggest that it may not be correct.

Keywords: Likelihood ratio test; Normal mixture

Document Type: Research article

Affiliations: 1: Biostatistics Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland 20892, U.S.A., Email: neal.jeffries@nih.gov

Publication date: 2003-12-01