Support drawing of cplot() effects for factors with more than two levels

[leeper]

Currently, an error is issued when trying to display the conditional effect(s) of a factor variable that has more than two levels in cplot():

> cplot(lm(y~x*fac,data=df), "x", "fac", what = "effect")
Error in cplot.lm(lm(y ~ x * fac, data = df), "x", "fac", what = "effect") : 
  Displaying on a factor variable with > 2 levels is not currently supported!

This could be fixed to display effect of each level/transition for the factor. margins() already does this fine, but it's a matter of figuring out how to draw it and return the data it appropriately.

[cschwem2er]

Hi,
first of all thank you so much for finally bringing margins to R. After seeing cplot() in the description and your example in this issue, I'm still not sure whether it is already possible to predict margins of a categorical across another continuous variable like so:
http://www.stata.com/stata14/i/mplot2.png
Will this be possible?

[leeper]

@methodds Not currently. What package do you use to for multinomial or ordered response variables? Is it with MASS::polr()? If so, this should be feasible but it is not currently supported (cplot() currently only has methods for "lm", "glm", and "loess" models).

[cschwem2er]

I'm not sure whether you perhaps misunderstood what I'm looking for. The different lines in the plot above refer to categories of an explanatory variable. The dependent variable still reflects probabilities from a logit model.

[leeper]

@methodds Ah, sorry. I thought the graph was predicted probabilities for a multiple outcome regression (ordered/multinomial logit) model.

If you just want to show predictions, that's currently possible. It just requires running cplot() for every level of the factor, like this (specifying draw = "add" for subsequent plots), so that a new predicted probability line is added for each separate factor level:

library("datasets")
library("stats")
library("margins")
iris$long <- rbinom(nrow(iris), 1, 0.6)
x <- glm(long ~ Sepal.Width*Species, data = iris)
cplot(x, x = "Sepal.Width", data = iris[iris$Species == "setosa", ], ylim = c(0,1), col = "red", se.fill = NA, xlim = c(0,5))
cplot(x, x = "Sepal.Width", data = iris[iris$Species == "versicolor", ], draw = "add", col = "blue", se.fill = NA)
cplot(x, x = "Sepal.Width", data = iris[iris$Species == "virginica", ], draw = "add", col = "green", se.fill = NA)

It's not ideal, but should get you to what you need.

[cschwem2er]

just what I needed! thanks for the example :)

[leeper]

I've added this example to the documentation. I'm going to close this for now; may revisit later.