Plot Probability Distribution

Produces a plot of a probability distribution function with shaded areas. Useful for illustrating critical regions, p-values, or probability masses in statistics teaching materials.

Usage

plotProbDist(
  breaks,
  FUN,
  main = "",
  xlim = NULL,
  col = NULL,
  density = 7,
  ylab = "density",
  areaLabels = NULL,
  breakLabels = NULL,
  grid = FALSE,
  box = .useTheme,
  ...
)

Arguments

breaks

numeric vector of break points defining the boundaries between shaded areas. The first and last value define the plot range passed to shade(); interior values are the actual boundaries.

FUN

the distribution density function to plot, typically of the form function(x) dnorm(x, mean=0, sd=1).

main

main title for the plot. NULL or "" suppresses the title and reduces the top margin automatically.

xlim

numeric vector of length 2. x-axis limits passed to curve().

col

color(s) for the shaded areas. Recycled if shorter than the number of areas. Defaults to the "helsana" palette.

density

density of shading lines passed to shade(). Default is 7.

ylab

label for the y-axis. Default is "density".

areaLabels

controls labels placed in the centre of each shaded area. NULL (default) suppresses labels. TRUE uses LETTERS as default labels. A character vector sets explicit labels. A named list overrides individual arguments passed to boxedText() (e.g. list(cex = 3) or list(x = c(-2, 3), y = 0.1) for manual positioning).

breakLabels

controls labels placed on the x-axis at interior break points via mtext(). NULL (default) suppresses labels. TRUE uses LETTERS as default labels. A character vector sets explicit labels. A named list overrides individual arguments (e.g. list(cex = 1.8, font = 1)).

grid

controls background grid. FALSE (default) suppresses the grid. TRUE or .useTheme draws a grid according to the current theme. A named list overrides individual arguments passed to grid().

box

controls the plot box. .useTheme (default) uses the current theme setting. TRUE/FALSE forces the box on or off. A named list overrides individual arguments passed to box().

...

further graphical parameters passed to curve() and .applyParFromDots(), e.g. las, col.axis.

Value

NULL, invisibly.

See also

curve()

Other plot.distribution: plotFun(), shade()

Examples

# Normal distribution with two areas
plotProbDist(breaks = c(-10, -1, 12),
             FUN    = function(x) dnorm(x, mean = 2, sd = 2),
             main   = "Normal-Distribution N(2,2)",
             xlim   = c(-7, 10),
             col    = c("deeppink4", "skyblue3"),
             density = c(20, 7),
             breakLabels = TRUE)


# t-distribution with three areas
plotProbDist(breaks = c(-6, -2.3, 1.5, 6),
             FUN    = function(x) dt(x, df = 8),
             main   = "t-Distribution (df=8)",
             xlim   = c(-4, 4),
             col    = c("deeppink4", "skyblue3", "darkorange2"),
             density = c(20, 7),
             areaLabels = FALSE,
             breakLabels = list(text=c("A", "B")))


# Chi-square distribution
plotProbDist(breaks  = c(0, 15, 35),
             FUN     = function(x) dchisq(x, df = 8),
             main    = expression(chi^2~"-Distribution (df=8)"),
             xlim    = c(0, 30),
             col     = c("deeppink4", "skyblue3"),
             density = c(0, 20),
             breakLabels = list(text="B"))
#> Warning: is.na() applied to non-(list or vector) of type 'expression'