plot.icsOut: Distances Plot for an icsOut Object

In ICSOutlier: Outlier Detection Using Invariant Coordinate Selection

Distances Plot for an icsOut Object

Description

Distances plot for an icsOut object visualizing the separation of the outliers from the good data points.

Usage

## S4 method for signature 'icsOut,missing'
plot(x, pch.out = 16, pch.good = 4, col.out = 1, col.good = grey(0.5), 
     col.cut = 1, lwd.cut = 1, lty.cut = 1, xlab = "Observation Number", 
     ylab = "ICS distances", ...)

Arguments

x	object of class icsOut.
pch.out	ploting symbol for the outliers.
pch.good	plotting symbol for the ‘good’ data points.
col.out	color for the outliers.
col.good	color for the ‘good’ data points.
col.cut	color for cut-off line.
lwd.cut	lwd value for cut-off line.
lty.cut	lty value for cut-off line.
xlab	default x-axis label.
ylab	default y-axis label.
...	other arguments for plot

Details

For the figure the IC distances are plotted versus their index. The cut-off value for distances is given as a horizontal line and all observations above the line are considered as outliers.

Author(s)

Aurore Archimbaud and Klaus Nordhausen

See Also

icsOut-class and ics.outlier

Examples

    Z <- rmvnorm(1000, rep(0, 6))
    Z[1:20, 1] <- Z[1:20, 1] + 10
    A <- matrix(rnorm(36), ncol = 6)
    X <- tcrossprod(Z, A)
    icsX <- ics2(X)
    # For demonstation purposes mDist is small, should be larger for real data analysis
    icsXoutliers <- ics.outlier(icsX, mDist = 500)
    plot(icsXoutliers, col.out = 2)

ICSOutlier documentation built on Sept. 18, 2023, 5:07 p.m.