plot.cubinf: Diagnostic plots for an object of class "cubinf"

In robcbi: Conditionally Unbiased Bounded Influence Estimates

Diagnostic plots for an object of class "cubinf"

Description

Six plots are available: Residuals vs Fitted Values, Sqrt of abs(Residuals) vs Fitted Values, Response vs Fitted Values" and QQline of Residuals. r-f spread plot is not available and Cook's distances are not available for objects of class "cubinf".

Usage

## S3 method for class 'cubinf'
plot(x, residuals = NULL, smooths = FALSE, rugplot = FALSE, id.n = 0, 
  ask = TRUE, num=0,  ...)

Arguments

x	An object of class "cubinf"
residuals	The residuals to be used in the plots if not null.
smooths	Logical indicating if a smoother should be added to most plots.
rugplot	Logical indicating if a "rug" representation of the data should be added to the plot.
id.n	Number of points to be labelled in each plot, starting with the most extreme.
ask	If ask=TRUE, the function operates in interactive mode.
...	Optional arguments for par.
num	Integer between 0 and 6. If num>0, plot the num-th choice in the previous list in batch mode.

See Also

glm.fit, plot.default

Examples

 library(robcbi)
 data(Finney)
 Vol <- Finney$Vol; Rate <- Finney$Rate; Resp <- Finney$Resp
 lVol <-log(Vol); lRate <- log(Rate)
 z.cub <- glm(Resp~lVol+lRate,family=binomial,method="cubinf", ufact=3.2)
 summary(z.cub)
## Not run: 
 plot(z.cub, smooth=TRUE, ask=TRUE)

## End(Not run)

robcbi documentation built on Aug. 22, 2023, 1:06 a.m.