A Short Introduction to Causal Statistics
This course is a practical, beginner-friendly introduction to causal inference for data scientists and analysts who already know some statistics and Python. It starts from the fundamental problem of causal inference — the fact that we can never observe both potential outcomes for the same unit — and builds up to modern heterogeneous treatment effect estimation using meta-learners (S-, T-, and X-learners) and evaluation tools such as the Qini curve and AUUC. Students will use two of the most widely adopted open-source libraries, CausalML and EconML, and leave with the ability to design, fit, and evaluate uplift models on real-world data.
- Level: beginner
- Duration: 20 days
- 6 sections
- Price: $20
Topics: causal inference, potential outcomes, treatment effect, CATE, uplift modelling, meta-learners, S-learner, T-learner, X-learner, Qini curve, AUUC, CausalML, EconML, randomised experiment, observational study
What's inside
- A Short Introduction to Causal Statistics
- Chapter 1: Why Causation Is Not Correlation
- Chapter 2: Heterogeneous Treatment Effects and the CATE
- Chapter 3: Meta-Learners: S-, T-, and X-Learner
- Chapter 4: Evaluating Uplift Models: Qini and AUUC
- Chapter 5: Putting It All Together: A Full Causal Uplift Pipeline