Causal Inference - Regression Discontinuity Design

Description

Regression Discontinuity Design (RDD) is a quasi-experimental research method that leverages a naturally occurring discontinuity or threshold to estimate the causal effect of a treatment or intervention. This project is a deeper dive into RDD and helps you learn about the theory & concepts using a practical example.

Features

You will learn about

• How RDD works
• Potential use cases for implementing RDD
• Sharp RDD - best practices for handling linear and non-linear relationship
• Fuzzy RDD - computing Intention to Treat Effect (ITTE) & compliance rate

Getting Started

Prerequisites

• Basic knowledge of Causal Inference - Ladder of causation, causal diagrams, do-calculus, etc.
• Basic knowledge of Linear Regression

Implementation

• Execute RDD.ipynb to learn from a practical example

Medium article

You can learn more on this topic from my article here