ltsspcaRw {ltsspca}R Documentation

Reweighted LTS-SPCA

Description

the function that computes the reweighted LTS-SPCA

Usage

ltsspcaRw(x, obj, k = NULL, alpha = 0.5, co.sd = 0.25)

Arguments

x

the input data matrix

obj

initial LTS-SPCA object given by ltsspca function

k

dimension of the PC subspace; by default is NULL then k takes the value of kmax in the initial LTS-SPCA

alpha

the robust parameter which takes value between 0 to 0.5, default is 0.5

co.sd

cutoff value for score outlier weight, default is 0.25

Value

the object of class "ltsspcaRw" is returned

loadings

the sparse loading matrix estimated with reweighted LTS-SPCA

scores

the estimated score matrix

eigenvalues

the estimated eigenvalues

mu

the center estimate

rw.obj

the list that contains the results of sPCA_rSVD on the reduced data

od

the orthonal distances with respect to the initially estimated PC subspace with all the noisy variables removed

co.od

the cutoff value for the orthogonal distances

ws.od

if the observation is outlying in the orthgonal complement of the initially estimated PC subspace ws.od=0; otherwise ws.od=1

sc.wt

the score outlier weight, which is compared with 0.25 (by default) to flag score outliers

co.sd

the cutoff value for score outlier weight, default is 0.25

ws.sd

if the observation is outlying with the PC subspace ws.sd=0; otherwise ws.sd=1

sc.out

the retruned object when computing the score outlier weights


[Package ltsspca version 0.1.0 Index]