my.lasso.classify: Sparse Optimal Discriminant Clustering

Description Usage Arguments Value

Description

To obtain SODC components and other relevant predictions in SODC method.

Usage

1
my.lasso.classify(data, c, lambda1, lambda2, tol = 10^(-10), iter.max = 50)

Arguments

data

A numberic dataset matrix.

c

An integer scalar with the desired number of groups.

lambda1

L1 penalty parameter, if lambda1=-1, then odc.clust choose the optimal lambda2 automatically given lambda2.idx. Otherwise perfoms SODC clustering using given lambda1.

lambda2

L2 penalty parameter, if lambda2=-1, then odc.clust choose the optimal lambda2 automatically given lambda2.idx. Otherwise perfoms ODC clustering using given lambda2.

tol

A tolerance value indicating the degree of prediction error of W.

iter.max

the maximum number of iterations allowed.

Value

Z

The SODC component.It is an n by k-1 matrix where n is the number of observations.

varset

An array indicating the selected varialbes index numbers.

what

Predicted W using SODC method.

nvarselected

The number of selected varialbes by SODC.The smaller the value, the sparser.


SODC documentation built on May 2, 2019, 3:35 p.m.