About: Predictive discrimination and classification with mixed binary and continuous variables

Documentation

scienceplus.abes.fr version Bêta

scienceplus.abes.fr | explorer

À propos de : Predictive discrimination and classification with mixed binary and continuous variables Goto Sponge NotDistinct Permalink

Attributs	Valeurs
type	work Article
Is Part Of	http://hub.abes.fr/oup/periodical/biomet/1990/volume_77/issue_3/w
Subject	Mixture of binary and continuous variables Location model Discriminant analysis Predictive and estimative allocation rule MISCELLANEA Classification error rate
Title	Predictive discrimination and classification with mixed binary and continuous variables
has manifestation of work	http://hub.abes.fr/oup/periodical/biomet/1990/volume_77/issue_3/101093biomet773657/m/web http://hub.abes.fr/oup/periodical/biomet/1990/volume_77/issue_3/101093biomet773657/m/print
related by	http://hub.abes.fr/oup/periodical/biomet/1990/volume_77/issue_3/101093biomet773657/authorship/1
Author	VLACHONIKOLIS I. G.
Abstract	We derive the predictive allocation rule for classification of observations involving mixtures of binary and continuous variables. Our approach is based on the usual frequency distributions of the location model and vague prior distributions for the unknown parameters. The same approach is used to derive predictive rules when low-order linear models are imposed on the population parameters. We also present Monte Carlo results on the performance of the new predictive rule and its estimative competitor.
article type	other
publisher identifier	77.3.657
is part of this journal	Biometrika

Alternative Linked Data Documents: ODE Content Formats:

RDF

ODATA

Microdata