Generalized Poisson distribution

Hey PyMC developers,

For a recent project, I decided to use a Generalized Poisson distribution as it offers a more flexible likelihood for modeling count data than the standard Poisson. I defined a custom discrete distribution in PyMC3 to do this, and I’d love to share what I did with the community. If you’re interested in adding an implementation for the Generalized Poisson, I could open a PR or share my code with you all. I’m also thinking of writing a blog post to give a tutorial on how I defined a custom distribution in PyMC3, which could be useful for your docs. Let me know if you’d like to collaborate on any of this!



Not a developer but I’d love to read the blog post!