minimaxlda {RSSL}

Implements weighted likelihood estimation for LDA

Description

Implements weighted likelihood estimation for LDA

Usage

minimaxlda(a, w, u, iter)

Arguments

a

is the data set

w

is an indicator matrix for the K classes of a or, potentially, a weight matrix in which the fraction with which a sample belongs to a particular class is indicated

u

is a bunch of unlabeled data

iter

decides on the amount of time we spend on minimaxing the stuff

Value

m contains the means, p contains the class priors, iW contains the INVERTED within covariance matrix, uw returns the weights for the unlabeled data, i returns the number of iterations used