Learn R Programming

extremevalues (version 2.3.4)

outlierPlot: Plot results of outlierdetection

Description

This is a wrapper for two plot functions which can be used to analyse the results of outlier detection with the extremevalues package.

Usage

outlierPlot(y, L, mode="qq", ...)
qqFitPlot(y, L, title=NA, xlab=NA, ylab=NA, fat=FALSE)
plotMethodII(y, L, title=NA, xlab=NA, ylab=NA, fat=FALSE)

Arguments

y

A vector of values

L

The result of L <- getOutliers(y,...)

mode

Plot type. "qq" for Quantile-quantile plot with indicated outliers, "residual" for plot of fit residuals with indicated outliers (Method II only)

...

Optional arguments, to be transferred to qqFitPlot or plotMethodII (see below)

title

A custom title (must be a string)

xlab

A custom label for the x-axis (must be a string)

ylab

A custim label for the y-axis (must be a string)

fat

If TRUE, axis, fonts, labels, points and lines are thicker for export and publication

Author

Mark van der Loo, www.markvanderloo.eu

Details

Outliers are marked with a color or special symbol. If mode="qq": observed agains predicted y-values are plotted. Points between vertical lines were used in the fit. If L$method="Method I", horizontal lines indicate the limits below (above) which observations are outliers. mode="residuals" only works when L$Method="Method II". It generates a residual plot where points between two vertical lines were used in the fit. Horizontal lines indicate the computed confidence limits. The outermost points in the gray areas are outliers.

References

The file <your R directory>/R-<version>/library/extremevalues/extremevalues.pdf contains a worked example. It can also be downloaded from my website.

Examples

Run this code
y <- rlnorm(100)
y <- c(0.1*min(y),y,10*max(y))
K <- getOutliers(y,method="I",distribution="lognormal")
L <- getOutliers(y,method="II",distribution="lognormal")
par(mfrow=c(1,2))
outlierPlot(y,K,mode="qq")
outlierPlot(y,L,mode="residual")

Run the code above in your browser using DataLab