This repository contains Python-based implementations and illustrative examples of key quasi-experimental methods used in causal inference. These methods are especially useful when randomized controlled trials are not feasible, and researchers must rely on observational data to estimate causal effects.
Quasi-experimental methods help evaluate the impact of interventions or treatments while accounting for potential confounders. The repository includes code and explanations for the following methods:
- ✅ Difference-in-Differences (DiD)
- ✅ Interrupted Time Series (ITS)
- ✅ Regression Discontinuity Design (RDD)
- ✅ Instrumental Variables (IV)
- ✅ Propensity Score Matching (PSM)
Each method is implemented with simple, well-commented code to demonstrate how it works in practice, often using real or simulated data.
These methods are commonly used in fields like:
- Public policy evaluation
- Health economics
- Education research
- Epidemiology
- Social sciences