How do I create an interaction plot when doing a negative binomial regression?

Question

I have conducted a negative binomial regression and utilized the glm.nb function from the MASS-package. My model includes 2 interaction effects and I would therefore like to create interaction plots. I tried to use the interplot function from the interplot-package:

interplot(m = model, var1 = "aid1", var2 = "aid")**

interplot(m = model, var1 = "Adjusted.aid1", var2 = "Adjusted.aid2")**

However, I get the following error:

Error in (function (classes, fdef, mtable)  : 
  unable to find an inherited method for function ‘sim’ for signature ‘"negbin"’

Does anyone know how I can create interaction plots for my model? Can I use interplot? Any help would be highly appreciated.

Answer 1

You can try the marginaleffects package: (Disclaimer: I am the author.)

library(MASS)
library(marginaleffects)
mod <- glm.nb(gear ~ hp * drat, data = mtcars)

plot_cme(mod, effect = "hp", condition = "drat")