glmxdiag (version 1.0.0)

DFbeta: Visualize DFbetas

Description

Graphical visualization of DFbetas for each observation and chosen variables within input model.

Usage

DFbeta(object, variables, centered = TRUE, label.id, n.label.id, 
              xlab, ylab, main, pos, pch, cex, lcol, lwd, lty, ...)

Arguments

object

object of class 'influence', see Details.

variables

a vector of characters containing names of variables; if nothing is specified, DFbetas for all variables are plotted.

centered

logical, if TRUE the plot is centered on the coefficients of the full model and each point corresponds to the coefficient if the i-th observation is excluded. if FALSE, usual DFbetas centered on zero are plotted.

label.id

labels of observations, should be a vector with n elements.

n.label.id

number of observations with highest DFbetas to label in the plot.

xlab

title for the x axis.

ylab

title for the y axis.

main

an overall title for the plot.

pos

position of observations labels. Values 1, 2, 3, 4 respectively indicate below, left, above, right.

pch

type of points.

cex

size of points.

lcol

color of line.

lwd

width of line.

lty

type of line.

...

further arguments passed to plot

Value

Doesn't return a value, called for side effects.

Details

This function only works with objects given as output from influenceDiag.

DFbetas are calculated as B - B(-i) hence they are not standardized; if centered is set to TRUE then each plotted point corresponds to B(-i), otherwise to B - B(-i).

Examples

# NOT RUN {
data("mtcars")
mod <- glm(mpg ~ cyl + hp + carb, family = Gamma, data = mtcars)
inf <- influenceDiag(mod)
DFbeta(inf, label.id = rownames(mtcars))
DFbeta(inf, variables = 'cyl', label.id = rownames(mtcars), pos = 2)

# to visualize all of them in a single plot
par(mfrow = c(2, 2))
DFbeta(inf, label.id = rownames(mtcars))
# press <Enter> four times
par(mfrow = c(1, 1))

# }