scaleBreak: Scatterplot with a Scale Break

In vagnerfonseca/fifer: A Biostatisticians Toolbox for Various Activities, Including Plotting, Data Cleanup, and Data Analysis

Description

With skewed data, important relationships are often scaled so small because the graph attempts to capture the outliers. scaleBreak will plot two different scales on the same plot, shrinking the range where the outliers lay. The break in scale is indicated with a mark.

Usage

scaleBreak(x, y, axis = 2, breakpos = 1, plot.numbers = c(1, 2), ...)

Arguments

x	The x variables to be plotted.
y	The y variables to be plotted.
axis	Which axis should have two different scales. Currently, only implemented for "2," which is the y axis.
breakpos	At what point should the break occur? Defaults to 1.
plot.numbers	Which plots to include. Can be 1, 2, or c(1,2).
...	other arguments passed to plot

Author(s)

Dustin Fife

Examples

# generate correlated data
d = data.frame(mvrnorm(1000, mu=c(0,0), Sigma=matrix(c(1,.6,.6,1), nrow=2)))
names(d) = c("x","y")
## Skew Y
d$y = d$y^2
scaleBreak(d$x, d$y, breakpos=4, plot.numbers=1)
## add a lowess line
lines(lowess(d$x, d$y), col="red")
## add a fitted line
mod = glm(y~x, data=d, family=inverse.gaussian)
curve(predict(mod,data.frame(x=x),type="resp"),add=TRUE,col="blue")
## add second plot
scaleBreak(d$x, d$y, breakpos=4, plot.numbers=2)

vagnerfonseca/fifer documentation built on May 3, 2019, 4:06 p.m.