rrcov3way (version 0.1-10)

plot.tucker3: Plot a tucker3 object

Description

Different plots for the results of Tucker3 analysis, stored in a tucker3 object, see Details.

Usage

# S3 method for tucker3
plot(x, which = c("dd", "comp", "allcomp", "jbplot", 
    "tjplot", "all"), ask = (which == "all" && dev.interactive(TRUE)), id.n, ...)

Arguments

x

A tucker3 object

which

Which plot to select (see Details). Default is dd, Distance-distance plot.

ask

Generates all plots in interactive mode

id.n

Number of items to highlight

…

Other parameters to be passed to the lower level functions.

Details

Different plots for a tucker3 object will be produced. Use the parameter which to select which plot to produce:

dd

Distance-distance plot

comp

Paired component plot

jbplot

Joint biplot

tjplot

Trajectory plot

References

Kiers, H.A. (2000).Some procedures for displaying results from three-way methods. Journal of Chemometrics. 14(3): 151-170.

Kroonenberg, P.M. (1983).Three-mode principal component analysis: Theory and applications (Vol. 2), DSWO press.

Examples

# NOT RUN {
##
## The Bus data from package ThreeWay
##
data(Bus)       # from ThreeWay: 7x5x37, unfolded to 7x185
dim(Bus)

## Preprocessing
##
max.scale <- c(47, 10, 10, 15, 15)
maxBus <- rep(max.scale, 37)
BusN <- t(t(Bus)/maxBus)

## Use rarray to turn back to a 3-way array
##
X <- rarray(BusN, 7, 5, 37)
dimnames(X) <- list(dimnames(Bus)[[1]], c("L-Occ", "P-Occ", "Q-Occ", 
    "S-Occ", "R-Occ"), paste("V",1:37,sep=""))

## Perform classical Tucker3 (nor robust option and no 
##  ilr transformation for compositional data)
res <- Tucker3(X, robust=FALSE, ilr=FALSE)

## Distance-distance plot
plot(res, which="dd", main="Distance-distance plot")

## Paired component plot, mode A
plot(res, which="comp", main="Paired component plot (mode A)")

## Paired component plot, mode B
plot(res, which="comp", mode="B", main="Paired component plot (mode B)") 

## Joint biplot
plot(res, which="jbplot", main="Joint biplot") 

## Trajectory plot
plot(res, which="tjplot", main="Trajectory biplot") 


# }