Anthropometry (version 1.1)

xyplotPCA: PC scores for archetypes

Description

This function is a small modification of the generic xyplot function of the archetypes R package. It shows the scores for the principal components of all individuals jointly with the scores for the computed archetypes. This function is used to obtain the Figure 4 of the subsection 3.3 of Epifanio et al. (2013).

Value

• A device with the desired plot.

References

Epifanio, I., Vinue, G., and Alemany, S., (2013). Archetypal analysis: contributions for estimating boundary cases in multivariate accommodation problem, Computers & Industrial Engineering 64, 757--765.

See Also

archetypesBoundary, dataUSAF

Examples

#First,the USAF 1967 database is read and preprocessed (Zehner et al. (1993)).
m <- dataUSAF
#Variable selection:
sel <- c(48,40,39,33,34,36)
#Changing to inches: 
mpulg <- m[,sel] / (10 * 2.54)

#Data preprocessing:
preproc <- accommodation(mpulg,TRUE,0.95,TRUE) 

#Procedure and results shown in section 2.2.2 and section 3.1:
res <- archetypesBoundary(preproc$data,15,FALSE,3)

a3 <- archetypes::bestModel(res[[3]])
a7 <- archetypes::bestModel(res[[7]])

pznueva <- prcomp(preproc$data,scale=T,retx=T) 
#PCA scores for 3 archetypes:
p3 <- predict(pznueva, archetypes::parameters(a3)) 
#PCA scores for 7 archetypes:
p7 <- predict(pznueva,archetypes::parameters(a7))
#Figure 4 (a):
xyplotPCA(p3[,1:2],pznueva$x[,1:2],data.col=gray(0.7),atypes.col=1,atypes.pch=15)
#Figure 4 (b):
xyplotPCA(p7[,1:2],pznueva$x[,1:2],data.col=gray(0.7),atypes.col=1,atypes.pch=15)