Selection Bias Adjustment in the Pearlian Causal Framework

We'll cover Bareinboim, Tian, and Pearl's 2014 paper "Recovering from Selection Bias in Statistical and Causal Inference"

https://ftp.cs.ucla.edu/pub/stat_ser/r425.pdf

This paper covers theorems that can be reduced to common selection bias techniques like post-stratification weighting, and even machine-learning weighting approaches to make selection-bias corrected ML models.

We'll cover the basic theorems, and might go as far as doing a derivation or two (and even a demo) of weighting. Depends on how much time I'm able to spend for prep!

We're a very casual group, and generally take a discussion group format. Please come with question! If there are parts of the paper you didn't understand, just ask! Someone almost certainly has the same question, and we can all learn by going through the material in more detail.