CausalInferenceLab/Causal-Inference-with-Python

What it solves

This project makes a respected English-language textbook on causal inference accessible to Korean speakers. It removes the language barrier by providing a complete, carefully translated version of Matheus Facure's open book, so Korean-speaking data practitioners can learn concepts like randomized experiments, difference-in-differences, and meta-learners without needing to read the original English text.

How it works

The project is a web book built with Quarto from Jupyter notebooks. Each chapter is a notebook that contains both explanatory text (translated into Korean) and working Python code, letting readers follow along and run the examples themselves. The translation is done by a community of data analysts and scientists, with terminology aligned to a published Korean book and key terms shown in English on first mention. The repository includes a translation guide, a glossary, and automated GitHub Actions that rebuild the web book whenever changes are merged. It also runs a monthly check on the original book to detect updates and generate issues for the maintainers.

Who it’s for

It is for Korean-speaking data analysts, data scientists, researchers, and students who want to learn causal inference from scratch or deepen their practical skills. The content starts with basic concepts (potential outcomes, randomized experiments) and progresses to advanced methods such as synthetic control, orthogonal machine learning, and heterogeneous treatment effects. Since every chapter has executable Python code, it is also suitable for those who prefer hands-on learning.

Highlights

• Complete Korean translation of all 25 chapters and 5 appendices from the original open book.
• Every chapter includes Python code that can be run directly, making the material practical.
• Terminology is consistent with a popular Korean published book, and key terms are annotated with English originals.
• Community-driven: anyone can report issues, suggest improvements, or contribute via pull requests.
• Automated build and deployment to a free web book, with monthly tracking of upstream changes.