On the Path to Causal Inference
summary
A practical primer that claims causal effects can be estimated without an RCT by drawing DAGs, identifying backdoor paths and controlling the minimal covariates, illustrated with TripTease hotel data where on-site messaging increases conversion (after adjusting for triggers and traffic source) rather than a simple correlation. It shows how colliders and mediators mislead linear models, recommends use DAGs to identify minimal control set, and demonstrates tooling like statsmodels DoWhy and the apple pie package for validation.
AI-written summary · about
chapters (9)
- 0:00Causal inference basics
- 3:15Why causal inference matters
- 8:18Ladder of causation framework
- 16:49First causal diagram for Tripti
- 21:07Direct effects and backdoor paths
- 26:19Triggers and confounding variables
- 32:04Biases in causal models
- 34:56Key takeaways and resources
- 37:24Q&A
related talks
31mA Gentle Introduction To Causal Inference
43mA Practical Guide to Causality in Python (For The Perplexed)
29m🧠🧹 Causality - Mental Hygiene for Data Science
31mIntroduction to Causal Inference using pgmpy
24mAn introduction to Causal Inference with Python – making accurate estimates of cause and effect from
24mA Friendly Introduction To Causality
01h 01mThe Mathematics of Causal Inference, with Reflections on Machine Learning and the Logic of Science
31mPython and the Holy Grail of Causal Inference