GENERALIZATION OF THE MAHALANOBIS DISTANCE IN THE MIXED CASE

被引：32

作者：

BARHEN, A ^{[1
]}

DAUDIN, JJ ^{[1
]}

机构：

[1] INST NATL AGRON PARIS GRIGNON,F-75231 PARIS 05,FRANCE

来源：

JOURNAL OF MULTIVARIATE ANALYSIS | 1995年 / 53卷 / 02期

关键词：

DISCRIMINATION WITH DISCRETE AND CONTINUOUS VARIABLES; KULLBACK-LEIBLER DIVERGENCE; LOCATION MODEL;

D O I：

10.1006/jmva.1995.1040

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

The Mahalanobis distance is extended to the case where the variables are a mixture of discrete and continuous variables. The distributional properties are studied for the location model and asymptotical results are obtained for the general parametric case. The rate of convergence is evaluated through simulation. An example of these techniques is presented. (C) 1995 Academic Press, Inc.

引用

页码：332 / 342

页数：11

共 8 条

[1]

BARHEN A, 1994, THESIS U PARIS SUD O

[2] SELECTION OF VARIABLES IN MIXED-VARIABLE DISCRIMINANT-ANALYSIS [J].

DAUDIN, JJ .

BIOMETRICS, 1986, 42 (03) :473-481

[3] DISTRIBUTION OF QUADRATIC FORMS AND SOME APPLICATIONS [J].

GRAD, A ;

SOLOMON, H .

ANNALS OF MATHEMATICAL STATISTICS, 1955, 26 (03) :464-477

[4]

KRZANOWSKI WJ, 1983, BIOMETRIKA, V70, P235

[5]

KRZANOWSKI WJ, 1980, BIOMETRICS, V36, P486

[6]

KULLBACK S, 1959, INFORMATION THEORY S

[7]

Mardia K. V., 1988, MULTIVARIATE ANAL

[8]

RAO CR, 1973, LINEAR STATISTICAL I

← 1 →