ON THE MINIMAL DIMENSION OF SUFFICIENT STATISTICS

被引：2

作者：

BUTUROVIC, LJ

机构：

[1] Institute of Applied Mathematics and Electronics, Belgrade

来源：

IEEE TRANSACTIONS ON INFORMATION THEORY | 1992年 / 38卷 / 01期

关键词：

PATTERN RECOGNITION; SUFFICIENT STATISTICS; DIMENSION REDUCTION; BAYES ERROR PROBABILITY; MULTIVARIATE EXPONENTIAL FAMILIES;

D O I：

10.1109/18.108269

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

The minimal dimension of sufficient continuous statistics for a finite family of continuous probability density functions is determined. The relation between sufficient statistics and optimal statistics in the context of Pattern recognition systems is studied, and the application of the main result to the case of exponential families is discussed.

引用

页码：182 / 186

页数：5