ggDiagnose.lm: Diagnostic plot for lm and glm objects (ggplot based)

In benjaminleroy/ggDiagnose: Wrapper for Diagnostic Plots with ggplot2

Description

This function leverages code from the plot.lm function from the stats library. It allows for the same imput (except when related directly to 'par' functionality, which makes no sense to include). We also include "newer" functionality.

Usage

ggDiagnose.lm(x, which = c(1L:3L, 5L),
  caption = list("Residuals vs Fitted", "Normal Q-Q", "Scale-Location",
  "Cook's distance", "Residuals vs Leverage",
  expression("Cook's dist vs Leverage  " * h[ii]/(1 - h[ii]))),
  sub.caption = NULL, main = NULL, ..., id.n = 3,
  labels.id = factor(names(stats::residuals(x)), levels =
  names(stats::residuals(x))), qqline = TRUE, cook.levels = c(0.5, 1),
  show.plot = TRUE, return = FALSE, shape = 1, nrow = min(2,
  length(which)), smooth.color = "blue", dashed.color = c("red",
  "blue", NA, NA, "red", "black"))

Arguments

x	lm or glm object
which	which plots you'd like to create
caption	title per plot
sub.caption	caption for bottom of multiple plot visual (defaults to the formula of the model.)
main	title for the mulitple plot visual
...	extra attributes (currently not used)
id.n	id the n observations with largest residuals
labels.id	labels for all observations
qqline	logic for whether a qqline should be drawn (a line between the 25 and 75 quantiles in the Q-Q plot)
cook.levels	levels of cooks distance to visualize in the leverage vs standardized residual graphic
show.plot	logic to display the graphics (group of graphics in this case)
return	logic to return list of graphics and the data frame to make the majority of graphics
shape	shape of points (the default is 1, an open circle)
nrow	number of rows in the displayed graphic
smooth.color	color for smoothing lines
dashed.color	color for dashed line (a vector of length 6 is expected)

Details

This function can deal with lm and glm objects.

Value

depending on show.plot and return it will return the visualization of the graphics and/or a list of both the data frame used the make the majority of the graphic and a list of each individual graphic.

See Also

see dfCompile.lm for data creation.

Examples

lm.object <- lm(Sepal.Length ~., data = iris)

par(mfrow = c(2,3))
plot(lm.object, which = 1:6)

ggDiagnose.lm(lm.object, which = 1:6)

benjaminleroy/ggDiagnose documentation built on May 4, 2019, 3:07 a.m.