Reference Materials for Learning Causality

This article strikes the start of my new series on causality. In this article, I will share the reference materials that I find very useful through my journey in causal discovery and causal inference. Feel free to let me know if you have any comments or feedback. Let’s enjoy the causal journey together!

Courses

Starters

1. HarvardX PH559X — Causal Diagrams: Draw Your Assumption Before Your Conclusion: this course is taught by Miguel Hernán in Harvard and has a lot of examples to help understand the underlying assumptions of the causal graphs and how we use them. Highly recommend.
2. Introduction to Causal Discovery: this course is taught by Daniel Malinsky in a Bootcamp as well as in 2024 ACIC conference. The slides and videos both provide a comprehensive understanding of causal discovery and are great materials for starters to quickly know the field.

More Advanced

1. Machine Learning & Causal Inference: A Short Course
2. 6.S091: Causality
3. Targeted Learning Webinar Series
4. Awesome Causal Inference

Books

Starters

1. Pearl, J., & Mackenzie, D. (2018). The book of why: the new science of cause and effect. Basic books.: this book is a great introduction to people who doesn’t know causality is and want to have a deeper understanding of what causality means exactly. No technical details and very friendly to people who have no math or statistics background.
2. Causal Inference for the Brave and True: a great book that illustrates different causal inference methods clearly with examples. Highly recommend to starters who need to understand the technical details of different methods.
3. Hernán MA, Robins JM (2020). Causal Inference: What If. Boca Raton: Chapman & Hall/CRC.: if there is only one book that I can recommend, I would always go with this one.
4. Morgan, S. L., & Winship, C. (2015). Counterfactuals and causal inference. Cambridge University Press.

More Advanced

1. Pearl, J. (2009). Causality. Cambridge university press.
2. Laan, M. V. D., & Rose, S. (2018). Targeted learning in data science: causal inference for complex longitudinal studies.
3. Van der Laan, M. J., & Rose, S. (2011). Targeted learning (Vol. 1, №3). New York: Springer.

Tools

1. DAGitty: a very convenient tool that can (1) draw a causal graph; (2) check identifications; (3) obtain local Markov conditions; (4) integrate into dowhy.