Skip to main content

Tools to implement regression discontinuity designs in Python.

Project description


rdd is a set of tools for implementing regression discontinuity designs in Python. At present, it only allows for inputs that are pandas Series or DataFrames. Check out the tutorial here for a guide to using this package.


To install using pip, run pip install rdd. See here for more information and to download from PyPI. Otherwise, you can download this repository.

Current Features:

  • Computes the Imbens-Kalyanaraman optimal bandwidth (see this article for details)
  • Creates a dataset containing only observations within a given bandwidth
  • Generates a reduced pandas DataFrame for graphical analysis (such as binned scatter plots) when the number of observations is large
  • Implements a sharp RDD
    • With or without user supplied equation and/or controls
    • Via OLS or WLS

Features to Come:

  • Tutorial on how to check rdd's statistical assumptions in Python, such as using:
    • Continuity plots
    • Density plots
    • Tests for discontinuities
    • Checking for balance and testing for random assignment (though not a requirement, it can still be a useful check)
  • McCrary tests
  • Implementation of Fuzzy RDD

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

rdd-0.0.3.tar.gz (4.4 kB view hashes)

Uploaded source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page