boxplot.BranchGLMVI.boot: Box Plot Method for BranchGLMVI.boot Objects

In BranchGLM: Efficient Best Subset Selection for GLMs via Branch and Bound Algorithms

Box Plot Method for BranchGLMVI.boot Objects

Description

Creates box-and-whisker plots of approximate null distributions for the modified variable importance values.

Usage

## S3 method for class 'BranchGLMVI.boot'
boxplot(
  x,
  which = "all",
  linecol = "red",
  linelwd = 2,
  horizontal = TRUE,
  lim = NULL,
  show.names = TRUE,
  lab = "Modified Variable Importance",
  main = NULL,
  las = ifelse(horizontal, 1, 2),
  ...
)

Arguments

x	a BranchGLMVI.boot object.
which	which approximate null distributions to plot, can use a numeric vector of indices, a character vector of names, or "all" for all variables. The default is to create box-and-whisker plots for each set of variables that are not kept in each model.
linecol	the color of the line which indicates the observed modified variable importance values.
linelwd	the width of the line which indicates the observed modified variable importance values.
horizontal	a logical value indicating if the boxplots should be horizontal.
lim	a numeric vector of length 2, giving the coordinates range.
show.names	set to TRUE or FALSE to override the defaults on whether an axis label is printed for each group.
lab	a label for the axis corresponding to the modified variable importance values.
main	a main title for the plot.
las	the style of axis labels, see more at par.
...	further arguments passed to boxplot.default.

Value

This only produces a plot, nothing is returned.

See Also

hist.BranchGLMVI.boot

Examples

Data <- iris
Fit <- BranchGLM(Sepal.Length ~ ., data = Data, family = "gaussian", link = "identity")

# Doing branch and bound selection 
VS <- VariableSelection(Fit, type = "branch and bound", metric = "BIC", 
showprogress = FALSE)

# Getting approximate null distributions
set.seed(40174)
myBoot <- VariableImportance.boot(VS, showprogress =  FALSE)

# Plotting boxplots of selected sets of variables
oldmar <- par("mar")
par(mar = c(4, 6, 3, 1) + 0.1)
boxplot(myBoot, las = 1)
par(mar = oldmar)

# Plotting boxplots of selected sets of variables
boxplot(myBoot, las = 1, cex.axis = 0.55)

BranchGLM documentation built on Sept. 28, 2024, 9:07 a.m.