Causal Effects via DAGs | How to Handle Unobserved Confounders

🤝 Want your team maximizing Claude? I run 1:1 and team AI workshops for companies doing $1M+ per year: https://aibuilder.academy/yt/ASU5HG5EqTM This is the 4th video in a series on causal effects. In the last video, we saw that we could evaluate any causal effect for a Markovian causal model. However, the question remained of how to handle models that are not Markovian. In this video, we start to answer this question via two quick-and-easy graphical criteria for evaluating causal effects. Series Playlist: https://www.youtube.com/playlist?list=PLz-ep5RbHosVVTz9HEzpI4d6xpWsc8rOa Blog: https://medium.com/towards-data-science/causal-effects-via-dags-801df31da794?sk=aa0947ca29e23fb3c1612e40deac38cf Resources: - An Introduction to Causal Inference by Judea Pearl: https://www.degruyter.com/document/doi/10.2202/1557-4679.1203/html - On Identifying Causal Effects by Tian & Shiptser: https://faculty.sites.iastate.edu/jtian/files/inline-files/tian-shpitser-2009.pdf Introduction - 0:00 Identifiability - 0:28 Markovian Models - 2:12 Unobserved Confounders - 3:19 Back & Front Door Criteria - 4:18 Back Door Path - 4:44 Blocking - 5:22 Back Door Criterion - 7:27 Front Door Criterion - 9:14