On the performance of the linear discriminant function for spherical distributions
From MaRDI portal
The purpose of this paper is to examine the performance of Fisher's linear discriminant rule for general spherical distributions. The author provides upper and lower bounds for the probabilities of correct classification which improve the bounds given by \textit{R. M. Haralick} [Pattern discrimination using ellipsoidally symmetric multivariate density functions. Pattern Recognition 9, 89-94 (1977)].
Recommendations
- scientific article; zbMATH DE number 3949519
- Optimal discriminant functions for normal populations
- The bias of Fisher's linear discriminant function when the variancies are not equal
- On the consistency properties of linear and quadratic discriminant analyses
- A lower bound on the performance of the quadratic discriminant function
Cites work
- A BIVARIATE GENERALIZATION OF STUDENT'S t-DISTRIBUTION, WITH TABLES FOR CERTAIN SPECIAL CASES
- Distance and decision rules
- scientific article; zbMATH DE number 3923895 (Why is no real title available?)
- scientific article; zbMATH DE number 3224150 (Why is no real title available?)
- scientific article; zbMATH DE number 3278469 (Why is no real title available?)
- scientific article; zbMATH DE number 3348541 (Why is no real title available?)
- scientific article; zbMATH DE number 3354336 (Why is no real title available?)
- scientific article; zbMATH DE number 3393603 (Why is no real title available?)
- scientific article; zbMATH DE number 3081880 (Why is no real title available?)
- On the theory of elliptically contoured distributions
- Pattern discrimination using ellipsoidally symmetric multivariate density functions
- Quadratic discriminant functions in pattern recognition (Corresp.)
- Quadratic forms in spherical random variables: Generalized noncentral x2 distribution
- Representations of the central and non-central t distributions
- Robust tests for spherical symmetry
- Some multivariate density functions of products of Gaussian variates
- Statistical classification with quadratic forms
- THE USE OF THE HANKEL TRANSFORM IN STATISTICS
Cited in
(12)- Minimum distance discrimination rules and success rates for elliptical normal mixtures
- Two LDF characterizations of the normal as a spherical distribution
- On the performance of a discriminant function
- The bias of Fisher's linear discriminant function when the variancies are not equal
- scientific article; zbMATH DE number 5717583 (Why is no real title available?)
- scientific article; zbMATH DE number 3949519 (Why is no real title available?)
- scientific article; zbMATH DE number 475217 (Why is no real title available?)
- scientific article; zbMATH DE number 572287 (Why is no real title available?)
- scientific article; zbMATH DE number 1098848 (Why is no real title available?)
- The Kotz-type distribution with applications
- A preliminary test in classification and probabilities of misclassification
- Discrimination of observations into one of two elliptic populations based on monotone training samples
This page was built for publication: On the performance of the linear discriminant function for spherical distributions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1822173)