Learn R Programming

JWileymisc (version 1.4.1)

plot.residualDiagnostics: Plot Residual Diagnostics Default Method

Description

This function creates a number of diagnostic plots from residuals. It is a default method.

Usage

# S3 method for residualDiagnostics
plot(x, y, plot = TRUE, ask = TRUE, ncol, ...)

Value

a list including plots of the residuals, residuals versus fitted values

Arguments

x

A residualDiagnostics class object.

y

Included to match the generic. Not used.

plot

A logical value whether or not to plot the results or simply return the graphical objects.

ask

A logical whether to ask before changing plots. Only applies to interactive environments.

ncol

The number of columns to use for plots. Missing by default which means individual plots are created. If specified, plots are put together in a grid.

...

Included to match the generic. Not used.

Examples

Run this code
testm <- stats::lm(mpg ~ hp * factor(cyl), data = mtcars)
testm <- stats::lm(mpg ~ factor(cyl), data = mtcars)

md <- residualDiagnostics(testm, ev.perc = .1)

plot(md, plot = FALSE)$ResFittedPlot
plot(md, ncol = 2)

## clean up
rm(testm, md)

Run the code above in your browser using DataLab