R: Plot observation weights from a robust multivariate linear...
plot.robmlm
R Documentation
Plot observation weights from a robust multivariate linear models
Description
Creates an index plot of the observation weights assigned in the
last iteration of robmlm. Observations with low weights
have large residual squared distances and are potential multivariate
outliers with respect to the fitted model.
Usage
## S3 method for class 'robmlm'
plot(x, labels,
id.weight = 0.7, id.pos = 4,
pch = 19,
col = palette()[1], cex = par("cex"),
segments = FALSE,
xlab = "Case index", ylab = "Weight in robust MANOVA", ...)
Arguments
x
A "robmlm" object
labels
Observation labels; if not specified,uses rownames from the original data
id.weight
Threshold for identifying obsrevations with small weights
id.pos
Position of observation label relative to the point
pch
Point symbol(s); can be a vector of length equal to the number of observations in the data frame
col
Point color(s)
cex
Point character size(s)
segments
logical; if TRUE, draw line segments from 1.o down to the point
xlab
x axis label
ylab
y axis label
...
other arguments passed to plot
Value
Returns invisibly the weights for the observations labeled in the plot