7
$\begingroup$

Over the past 15 years there has been progress in adapting machine learning methods for causal inference. For example: targeted learning, double machine learning, causal trees.

Is there a textbook that covers the current range of techniques? I haven't seen anything on Amazon, perhaps there are texts available on other sites? Or will be published soon?

$\endgroup$
3
  • $\begingroup$ ML doesn’t come to mind when thinking of causal inference $\endgroup$
    – Aksakal
    Oct 19 at 21:12
  • 3
    $\begingroup$ @Aksakal Agreed, but lately there's been some work in this area. ML algorithms have some advantages over traditional parametric models when estimating average treatment effects for complex, high dimensional data. The tricky part is finding unbiased estimates plus confidence intervals. $\endgroup$
    – RobertF
    Oct 20 at 2:32
  • $\begingroup$ Your question is about recommended books for ML and causal inference. Useful suggestions was already given. About causal inference and non recommended econometrics books read here stats.stackexchange.com/questions/477705/… $\endgroup$
    – markowitz
    Nov 14 at 12:02
5
$\begingroup$

I follow this area pretty closely, but I think this subfield is so new no textbook exists (yet).

However, there are some course videos that are fairly good:

  1. Machine Learning & Causal Inference: A Short Course at Stanford (accompanying tutorial)
  2. Summer Institute in Machine Learning in Economics (MLESI21) at University of Chicago

There is also a nice survey paper: "Machine learning methods that economists should know about" by Susan Athey, Guido Imbens in the Annual Review of Economics (link to draft)

$\endgroup$
1
  • 1
    $\begingroup$ Thanks dimitriy. Susan Athey's name comes up quite a bit when I google "machine learning causal inference" $\endgroup$
    – RobertF
    Oct 20 at 2:34
7
$\begingroup$

Since you are looking for more focused topics on causal inference

  1. Targeted Learning:
    This book, Targeted Learning: Causal Inference for Observational and Experimental Data

  2. Double Machine Learning

  3. Causal Trees

This link provides a general introduction of both double machine learning and causal trees



Other causal inference books and reference papers you may be interested in.
Below flowchart can help you out based on your preference Causal inference based on preference

BOOKS

  • Elements of Causal Inference: Foundations and Learning Algorithms :
    Most ideal for people in machine learning background. This book covers modern causal discovery (structure learning), connects causal inference to machine learning, causal discovery, foundation in structural causal models (SCMs).

  • Causal Inference: What If
    This book covers the concept known as “positivity” (or “overlap”) which helps to estimate causal effects, how to estimate the outcome models and propensity score models which are used to estimate causal effects in practice, different kinds of estimators that can be used in practice.

  • Causal Inference in Statistics: A Primer
    This book provides quick introduction to causal inference and SCMs.

  • Causality: Models, Reasoning and Inference
    Go for this book if you want to become expert in SCMs.

  • Counterfactuals and Causal Inference: Methods and Principles for Social Research
    This book cover topics that come from the causal inference literatures in economics, computer science, and across the social sciences with real examples. This book provides a great combination and comparison of the potential outcomes and graphical causal models perspectives. They thoroughly cover 3 different classes of conditioning-based estimators of causal effects, giving each their own chapter: matching, regression, and inverse probability weighting. They also have dedicated chapters to instrumental variables and the frontdoor adjustment.

  • Mostly Harmless Econometrics
    This book focuses on econometrics and places its emphasis on quasi-experimental methods such as instrumental variables, differences-in-differences, and regression discontinuity designs.

  • Causal Inference: The Mixtape
    This book covers recent methods like synthetic controls

  • Explanation in Causal Inference: Methods for Mediation and Interaction (VanderWeele, 2015) - This book is the go-to for becoming expert in mediation and interaction and is not very technical.

  • Observation and Experiment: An Introduction to Causal Inference (Rosenbaum, 2017)
    The concepts of causal inference, with reasonable precision, but with a minimum of technical material.

  • Actual Causality (Halpern, 2016) - This book captures what it actually means for X to cause Y helps to understand what it means for something to be a cause of something else.

  • Design of Observational Studies (2010)

  • Causation, Prediction, and Search (Spirtes et al., 2001)

  • Observation and Experiment: An Introduction to Causal Inference

  • Advanced Data Analysis from an Elementary Point of View

  • Targeted Learning (van der Laan & Rose, 2011)

Research papers

$\endgroup$
4
  • 1
    $\begingroup$ @robertf: I understand you were looking for books on causal inference, have also added focused information on double machine learning, causal trees and targeted learning. Hope the information provides is helpful. kindly consider to upvote & accept answer $\endgroup$ Oct 20 at 20:16
  • 1
    $\begingroup$ I just saw "I hate them" on the flow diagram and burst out laughing. $\endgroup$
    – Galen
    Nov 3 at 21:13
  • 3
    $\begingroup$ Nice post, but should probably make clearer that the flowchart comes from the preceding link [bradyneal.com/which-causal-inference-book] $\endgroup$ Nov 4 at 2:57
  • 3
    $\begingroup$ This is useful, but you should acknowledge the blog page bradyneal.com/which-causal-inference-book because most of your text about the books (+ the flowchart) is derived from this page. $\endgroup$ Nov 4 at 13:36
1
$\begingroup$

As dimitriy states, there isn't a singular textbook yet (or at least that I am aware of). However, there are a few textbook materials you can piece together to cover the topics you mentioned.

  1. Targeted Learning in Data Science covers super learner (which is a generalized stacking algorithm you would almost always want to use in practice), and targeted maximum likelihood estimation (with a bunch of variations of it). I think this one will be preferred over the other targeted learning book since the one linked above covers the machine learning parts a bit better
  2. Chapter 18 of Hernan and Robins covers double machine learning.

Unfortunately, I don't have a recommendation for causal trees

$\endgroup$
11
  • $\begingroup$ Beware that the super learning can be dominated by a single "machine" that is grossly overfitted, making the super learner ensemble highly overfitted. $\endgroup$ Oct 20 at 12:30
  • $\begingroup$ Do you have a reference that demonstrates that? My impression is that the cross-validation decreases that (more so than alternative approaches) $\endgroup$
    – pzivich
    Oct 20 at 12:37
  • $\begingroup$ I just have an example analysis on a 40,000 patient clinical trial. Cross-validation reveals but doesn't fix that. The super learner was fooled. One of the learners was a random forest-like method that was overfitted to a degree I've never seen before. $\endgroup$ Oct 20 at 12:42
  • $\begingroup$ @FrankHarrell A good topic for another question, I'd like to see this study. I'm curious if it's possible, even with cross-validation, to overfit a model to given training & test datasets? $\endgroup$
    – RobertF
    Oct 20 at 16:10
  • 1
    $\begingroup$ No, DML requires sample splitting. Chernozhukov et al. (2017) estimates the ATE using the augmented inverse probability weighting (AIPW) estimator with single cross-fitting. See stats.stackexchange.com/a/482498/247479 The 3rd model doesn't happen (that's almost like TMLE, but not quite). But influence curves are underlying all of it since they are used to show that the estimator is semiparametric efficient, and the variance (particularly with DML) is estimated from the influence curves. For how AIPW works, I would read Funk et al. (2011) "Doubly Robust Estimation of Causal Effects". $\endgroup$
    – pzivich
    Oct 21 at 14:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.