Last active
September 28, 2019 07:11
-
-
Save statwonk/0bff890bbdc78fabd28db536f88093a3 to your computer and use it in GitHub Desktop.
A gist that shows how a rootogram helps find that the zero-inflated negative binomial was the data generating mechanism.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# install.packages("countreg", repos="http://R-Forge.R-project.org") | |
# https://www.fromthebottomoftheheap.net/2016/06/07/rootograms/ | |
# https://channel9.msdn.com/Events/useR-international-R-User-conferences/useR-International-R-User-2017-Conference/countreg-Tools-for-count-data-regression | |
library(countreg) | |
rzinbinom(3e3, size = 4, mu = 20, pi = 0.05) -> x | |
table(x) | |
hist(x, col = "orange") | |
rootogram(glm(x ~ 1, family = "poisson")) # zeros under fit | |
rootogram(zeroinfl(x ~ 1, dist = "poisson")) # low counts under fit, high counts over fit | |
rootogram(zeroinfl(x ~ 1, dist = "negbin")) # zero fit well, low counts fit well, high counts fit well. | |
sort(ls("package:countreg")) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment