effectplot {boolean}R Documentation

Visually Assess Nonlinearity in Effects of Explanatory Variables

Description

This function plots the change in probability of the outcome in one explanatory variable as a function of a change in another explanatory variable.

Usage

effectplot(z.out, variables, delta = 1, CI = 95, truehist = TRUE,
                       legend = TRUE, polygon = FALSE, ...)

Arguments

z.out An object produced by Zelig.
variables A character vector of length two. The first indicates which explanatory variable should constitute the x-axis of the plot. The second indicates which explanatory variable should be perturbed by delta
delta The amount by which to perturb the sample values of the second element of variable.
CI Confidence interval for (the change in) the probability of the outcome. Set this equal to any negative number to suppress confidence intervals.
truehist If TRUE, a histogram for variable is included in the plot.
legend If TRUE, a legend is included in the plot.
polygon If TRUE, then the confidence intervals as filled as ploygons.
... Further arguments passed to plot or setx.

Value

Nothing is returned.

Author(s)

Bear F. Braumoeller, Harvard University, braumoeller.1@polisci.osu.edu,
Ben Goodrich, Harvard University, goodrich@fas.harvard.edu, and
Jacob Kline, Harvard University, jkline@fas.harvard.edu

References

Braumoeller, Bear F. (2003) "Causal Complexity and the Study of Politics." Political Analysis 11(3): 209-233.


[Package boolean version 2.0-2 Index]