A Practical Guide to Causality in Python (For The Perplexed)
summary
Practical steps and code in Python: builds causal graphs, encodes assumptions, uses front-door/back-door/do-calculus to translate queries into estimable expressions, and runs estimators from dowhy and econml (repo included) on synthetic and churn-prevention data. Shows how to detect colliders, forks and chains, runs refutation tests, and reports an estimated interventional effect that matches the known ground truth in the example.
AI-written summary · about
chapters (18)
- 0:00Causality in Python intro
- 1:16Santino the planning chimp
- 4:26Counterfactual reasoning
- 8:05Correlation vs causation
- 10:35Limitations of traditional machine learning
- 12:24What is causality
- 15:34Beyond Wikipedia
- 19:51Pearl’s interventional definition
- 20:36Ladder of causation
- 24:12Python practical tools
- 25:16Graphical models
- 27:48Chains forks colliders
- 30:58Feature selection fails
- 33:29Simpson’s Paradox
- 36:01Four steps of causal inference
- 37:34DoWhy and EconML
- 40:42Validating causal models
- 42:06Subgroup effects closing
related talks
01h 01mThe Mathematics of Causal Inference, with Reflections on Machine Learning and the Logic of Science
29m🧠🧹 Causality - Mental Hygiene for Data Science
41mOn the Path to Causal Inference
31mIntroduction to Causal Inference using pgmpy
33mCasual Notes about Causal Stuff
24mAn introduction to Causal Inference with Python – making accurate estimates of cause and effect from
45mCausality and Statistics, a Love Not So Impossible After All
01h 06mThe New Science of Cause and Effect