Regression Discontinuity: add ability to deal with fuzzy RD

[drbenvincent]

• Make it clear in the docs that RegressionDiscontinuity deals with sharp RD designs [closed by Update docs to make clear sharp regression discontinuity design #437]
• Add a new FuzzyRegressionDiscontinuity class with docstrings
• Add an example notebook
• Add to README.md (GitHub)
• Add to index.rst (readthedocs)

• Imbens, G. W., & Lemieux, T. (2008). Regression discontinuity designs: A guide to practice. Journal of econometrics, 142(2), 615-635.
• https://github.com/rdpackages/rdrobust / https://rdpackages.github.io
• https://clas.ucdenver.edu/marcelo-perraillon/sites/default/files/attached-files/regression_discontinuity_designs.pdf#:~:text=Sharp%20RDD%3A%20Assignment%20or%20running,doesn't%20completeley%20determines%20treatment (Link from @JacobElder in Support for fuzzy vs. sharp regression discontinuity? #333)

[HPCurtis]

What’s reason for removal of the request?

[drbenvincent]

Think I was just tidying up labels. Can't quite remember. But the issue is still open 😄

[HPCurtis]

Definitely interested in this giving this a go. Do you have any particular Bayesian fuzzy discontinuity implementation in mind? Various forms with a quick google scholar look at papers.

[drbenvincent]

That would be super cool 😎

I don't have a strong opinion at this point. But the preference would be for the API and general approach to be similar, if possible.

I'll maybe take a quick look, but feel free to use this issue to outline and talk through possible implementations.

[HPCurtis]

Not fully fleshed out by any means see notebook but this is functioning PyMC model on the simplest fuzzy regression discontinuity model. Basically a 2 stage regression model. Seems like the best starting approach. Let me know if happy with this generally and can start up proper with following the API and such.