Asymptotic distribution of Matusita's distance : application to the location model
The extension of Mahalanobis's distance to the case of a mixture of discrete and continuous variables is useful in many applications of discriminant analysis. In this note, the asymptotic chisquared distribution of the estimator of Matusita's distance under the null hypothesis is obtained and the rate of convergence is evaluated through simulations in the case of the location model.
Saved in:
Main Authors: | , |
---|---|
Format: | article biblioteca |
Language: | eng |
Subjects: | U10 - Informatique, mathématiques et statistiques, analyse de données, méthode statistique, modèle mathématique, simulation, biométrie, http://aims.fao.org/aos/agrovoc/c_15962, http://aims.fao.org/aos/agrovoc/c_7377, http://aims.fao.org/aos/agrovoc/c_24199, http://aims.fao.org/aos/agrovoc/c_5209, http://aims.fao.org/aos/agrovoc/c_927, |
Online Access: | http://agritrop.cirad.fr/390564/ http://agritrop.cirad.fr/390564/1/390564.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | The extension of Mahalanobis's distance to the case of a mixture of discrete and continuous variables is useful in many applications of discriminant analysis. In this note, the asymptotic chisquared distribution of the estimator of Matusita's distance under the null hypothesis is obtained and the rate of convergence is evaluated through simulations in the case of the location model. |
---|