Causality resources

4 minute read


Good resources matter, a lot.

The need for speed good resources

One of the most important steps before starting to study a topic is to collect some good resources, Scott H. Young argues in his 2019 book Ultralearning.

As a side note, I might—I mean, I will do this for sure—litter my posts with book references (actually, I am even planning to share the books I am reading with you in the near future).

But let’s not permit ourselves to be distracted, let’s get to the business!

As I agree with Scott, I invested some time to compile a list of materials, including lectures, books, MOOCs (Massive Open Online Courses). I have already used some of these, but at least I familiarized myself with it enough to deem each and every item valuable. Let’s get to the details, shall we?

Causality resources


  • Judea Pearl and Dana Mackenzie, The Book of Why: a gentle non-technical introduction into the topic with a lot of examples; definitely worth reading!
  • Judea Pearl, Causality (2009) : this is THE classic textbook on the topic. It is not a very easy read, but it contains a vast amount of theoretical background. Nonetheless, a lot of times it refers to other papers for proofs, etc. - I found this reasonable but inconvenient. Spoiler: one of the goals of this blog is to get through this book and explain the concepts to develop a better understanding.
  • Bernhard Schölkopf, Dominik Janzing, and Jonas Peters, Elements of Causal Inference: a recent open-access book that is keen on the theory of causal inference.
  • Brady Neal, Introduction to Causal Inference from a Machine Learning Perspective : this is the lecture notes for the online course Introduction to Causal Inference (listed in the MOOCs and online lectures section). Due to being lecture notes, I find this book to be more clearly cut, more structured, but it covers less material. Unfortunately, as of today, the book is still in progress-the most interesting (and advanced) chapters are still on their way.
  • Miguel A. Hernán, James M. Robins, Causal Inference: What if: this open-access book is mainly aimed at social scientists and medical researchers (in the sense that the authors mainly discuss causality with concepts widely-used in those domains). Nevertheless, graphs (the choice of weapon for computer scientists) are also discussed in this very detailed book.

MOOCs and online lectures

  • Probabilistic Graphical Models Specialization: this three-course feast presented by the famouse Daphne Koller provides you with the basics of traditional methods in graphical models. Although not causality-specific in its entirety, it lays the foundations, so definitely worth attending-it even provides programming assigments so you can get familiar with the methods. Aaaand it’s totally free (if you are satisfied without a Coursera certificate).
  • Introduction to Causal Inference: a causal inference course covering the fundamental topics. As it has the lecture notes released (see Books), you should be well-equipped to take the challenge to become a causality expert.
  • Bayesian Statistics: From Concept to Data Analysis: an introductory course into Bayesian statistics. Besides covering the base (such as Bayes’ theorem, Bayesian inference), it also discusses more advances topics as uninformative and Jeffreys priors. As a bonus, we get some insight how to apply the concepts in R.
  • Jonas Peters’ Lectures on Causality Part 1, Part 2, Part 3, Part 4 from 2017, held at MIT: these are superb introductory lectures on the topics with a sense of humor and a lot of examples.
  • Online Causal Inference Seminar: the name tells everything. This interdisciplinary seminar will provide you the viewpoints of different scientific fields on causality. Let’s exploit this unique opportunity to avoid falling into the bias of a single field’s approach. The recordings are oalso available so you cannot make any excuse!


P.S.: if you find any error in the decks, please contact me