getLogLikeRatio-methods

getLogLikeRatio,LinearRule-method

getLogLikeRatio,QuadraticRule-method


 ~~ Methods for function <code>getLogLikeRatio</code>  ~~


methods

This package provides an implementation of Linear discriminant analysis and quadratic discriminant analysis that works fine in very high dimension (when there are many more variables than observations).

Robin Girard

VHDClassification

Discrimination/Classification in very high dimension with linear
and quadratic rules.

getLogLikeRatio-methods function

<dl> <dt><code>signature(object = "LinearRule")</code></dt><dd> Returns a list with NormalVector and CenterVector. The loglikelihood ratio on x can be evaluated by 
L(x)=1/2<normalvector,x-centervector>. </normalvector,x-centervector></dd> <dt><code>signature(object = "QuadraticRule")</code></dt><dd> returns a list with a NormalVector, CenterVector, FormVector (3 vectors) and a numeric constant Constant. The loglikelihood ratio on x can be evaluated by
L(x)=-1/2<diag(formvector)(x-centervector),x-centervector>+<normalvector,x-centervector> -Constant </normalvector,x-centervector></diag(formvector)(x-centervector),x-centervector></dd> </dl>

getLogLikeRatio-methods: Methods for Function getLogLikeRatio

Description

Arguments

Methods