Plot a scatter plot of a binary variable. xvar is the continuous independent variable and yvar is the dependent binary variable

BinaryYScatterPlot(frame, xvar, yvar, title, ..., se = FALSE,
  use_glm = TRUE)

Arguments

frame

data frame to get values from

xvar

name of the independent column in frame

yvar

name of the dependent (output or result to be modeled) column in frame

title

title to place on plot

...

no unnamed argument, added to force named binding of later arguments.

se

if TRUE, add error bars (defaults to FALSE). Ignored if useGLM is TRUE

use_glm

if TRUE, "smooths" with a one-variable logistic regression (defaults to TRUE)

Examples

set.seed(34903490) x = rnorm(50) y = 0.5*x^2 + 2*x + rnorm(length(x)) frm = data.frame(x=x,y=y,yC=y>=as.numeric(quantile(y,probs=0.8))) frm$absY <- abs(frm$y) frm$posY = frm$y > 0 frm$costX = 1 WVPlots::BinaryYScatterPlot(frm, "x", "posY", title="Example 'Probability of Y' Plot")