ppPlot: Probability Plots for various distributions

Description

ppPlot creates a Probability plot of the values in x including a line.

Usage

ppPlot(x, distribution,  confbounds = TRUE, alpha, probs, main, xlab, ylab, 
       xlim, ylim, border = "red", bounds.col = "black", bounds.lty = 1, 
       grid = TRUE, box = TRUE, stats = TRUE, start, ...)

Arguments

vector containing the sample for ppPlot.

distribution

character string specifying the distribution of x. The function ppPlot will support the following character strings for distribution:

beta
cauchy
chi-squared

confbounds

boolean value: TRUE if confidence bounds should be drawn (default value).

alpha

significance level for the confidence bounds, set on 0.05 by default.

probs

vector containing the percentages for the y axis. All the values need to be between 0 and 1 If probs is missing it will be calculated internally.

main

an overall title for the plot: see title.

xlab

a title for the x axis: title.

ylab

a title for the y axis: title.

xlim

vector giving the range of the x-axis.

ylim

vector giving the range of the y-axis.

border

numerical value or single character string giving the color of interpolation line. By default border is set to red.

bounds.col

numerical value or single character string giving the color of confidence bounds lines. By default bounds is set to black.

bounds.lty

numerical value giving the color of confidence bounds lines. By default bounds is set to 1.

grid

logical value, deciding whether a grid will be added to the current plot. By default grid is set to TRUE.

box

logical value, deciding whether a box is drawn to around the current plot. By default box is set to TRUE.

stats

logical value deciding whether ppPlot returns the values listed under value. By default stats is set to TRUE.

start

A named list giving the parameters to be fitted with initial values. Must be supplied for some distribution. (see Details)

...

further graphical parameters (see par).

Value

ppPlot returns a list containing the following: x - x coordinates y - y coordinates int - Intercept slope - slope

Details

Distribution fitting is deligated to function fitdistr of the R-package MASS. For computation of the confidence bounds the variance of the quantiles is estimated using the delta method, which implies estimation of observed Fisher Information matrix as well as the gradient of the CDF of the fitted distribution. Where possible, those values are replaced by their normal approximation.

Examples

Run this code

#set up the plotting window for 6 plots
par(mfrow = c(3,2))

#generate random data from weibull distribution
x = rweibull(20, 8, 2)

#Probability Plot for different distributions
ppPlot(x, "log-normal")
ppPlot(x, "normal")
ppPlot(x, "exponential", DB = TRUE)
ppPlot(x, "cauchy")
ppPlot(x, "weibull")
ppPlot(x, "logistic")