I have tried to get my head around the concept of TMLE, but most references seem to be written by people who despise being understood (or maybe I am just hebetudinous). I have tried to read the paper

Targeted Maximum Likelihood Estimation: A Gentle Introduction

which is not as gentle as the title implies, in my opinion. Is there a reference that explains this approach at a, say university/Masters level? Or is the concept of TMLE much more complex that cannot be explained at this level?

  • 2
    $\begingroup$ I share your view that we much need an understandable intro to TMLE. And realism when it comes to its computational burden. $\endgroup$ May 9, 2019 at 11:19

1 Answer 1


I have found this paper to be more approachable: https://academic.oup.com/aje/article/185/1/65/2662306

Are you familiar with inverse probability weighting and plug-in estimators (e.g. the g-formula)? TMLE is essentially a way to merge these two estimators into a single estimator. I have found it helpful to have a good understanding of those before trying to approach TMLE.

This paper does a good job as an introduction to the g-formula: https://www.ncbi.nlm.nih.gov/pubmed/21415029

For inverse probability weights, I would recommend (despite it being a tad dense): https://cdn1.sph.harvard.edu/wp-content/uploads/sites/1268/2014/11/Marginal_Structural_Models_and_Causal_Inference_in.11.pdf

This paper on double-robustness for missing data does not directly describe TMLE, but I found it to be useful for a conceptual understanding: https://statnav.files.wordpress.com/2017/10/doublerobustness-preprint.pdf


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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