Causal inference resources

This is a list of some causal inference resources, which I update from time to time. You can also check out my posts on causal inference and A/B testing.

Books:

Causal Inference: What if by Miguel Hernán and Jamie Robins: The most practical book I’ve read. Highly recommended.
Trustworthy Online Controlled Experiments : A Practical Guide to A/B Testing by Ron Kohavi, Diane Tang, and Ya Xu: Building on the authors’ decades of industry experience, this is pretty much the bible of online experiments, which is how causal inference is often done in practice.
Why: A Guide to Finding and Using Causes by Samantha Kleinberg: A high-level intro to the topic. I discussed highlights in Why you should stop worrying about deep learning and deepen your understanding of causality instead.
Causality, Probability, and Time by Samantha Kleinberg: More technical than Kleinberg’s other book. As the title suggests, the element of time is central to the methods presented in the book. However, I’m still unsure about the practicality of those methods on real data. See my post Diving deeper into causality: Pearl, Kleinberg, Hill, and untested assumptions for more details.
Causal Inference in Statistics: A Primer by Judea Pearl, Madelyn Glymour, Nicholas P. Jewell: A fairly accessible introduction to Judea Pearl’s work. I didn’t find it that practical, but I believe it helped me understand the graphical modelling parts of Causal Inference by Hernán and Robins.
Elements of Causal Inference: Foundations and Learning Algorithms by Jonas Peters, Dominik Janzing, and Bernhard Schölkopf: The name of the book is an obvious reference to the classic book The Elements of Statistical Learning by Trevor Hastie, Robert Tibshirani, and Jerome Friedman. Unfortunately, the Elements of Causal Inference isn’t as widely applicable as Hastie et al.’s book – it contains some interesting ideas, but it appears that algorithms for causal learning from data with minimal assumptions aren’t yet scalable enough for practical use. This will probably change in the future.
Mostly Harmless Econometrics by Joshua D. Angrist and Jörn-Steffen Pischke: I started reading this book on my Kindle and was put off by some formatting issues. It also seemed like a less-general version of Pearl’s work. I may get back to it one day.
Causality: Models, Reasoning, and Inference by Judea Pearl: I haven’t read it, and I doubt it’d be very practical given the opinions of people who have. But maybe I’ll get to it one day.
The Book of Why: The New Science of Cause and Effect by Judea Pearl and Dana Mackenzie: An accessible overview of the field, focusing on Pearl’s contributions, but with plenty of historical background. Worth reading to get excited about the causal revolution.
Causal Machine Learning by Robert Osazuwa Ness: Still a draft as of September 2022, but it looks promising.

Articles:

Does water kill? A call for less casual causal inferences by Miguel Hernán: A great demonstration of why talking about causality requires well-defined interventions.
The C-Word: Scientific Euphemisms Do Not Improve Causal Inference From Observational Data by Miguel Hernán: A high-level summary of causal inference and the need to be explicit about the causal goals of scientific studies.
The Environment and Disease: Association or Causation? by Austin Bradford Hill: A classic discussion of the Bradford Hill criteria for causation. Highly recommended, as this 1965 paper also foresaw the problems with the statistical significance cult.
Causal inference in statistics: An overview by Judea Pearl: A summary of Pearl’s work, which may be somewhat dated at this point (it’s from 2009). It’s still worth reading if you’re not ready to commit to reading his books.
Simpson’s Paradox: An Anatomy by Judea Pearl: An explanation of Simpson’s paradox and its relationship to causal inference. This paper is worth reading, though I found that further reading is required to better understand why causal modelling “solves” the paradox.
Guidelines for estimating causal effects in pragmatic randomized trials by Eleanor J. Murray, Sonja A. Swanson, and Miguel A. Hernán. Once you get over the terminology gap, you see how these guidelines apply to any field where experiments don’t always go as planned.

Courses:

Causal Diagrams: Draw Your Assumptions Before Your Conclusions. A high-level introduction to causal diagrams by Miguel Hernán. Highly recommended for those who want to get a conceptual overview of how causal diagrams work and why they’re useful.
A/B Testing by Google: Online Experiment Design and Analysis. Experimentation is key to causal inference, with the online world offering an accessible ground for running experiments. This short course is worth doing if you’re involved in online experiments in any way.