Accelerated Foundation in Applied Causal Inference

Econometrics and data science have long lived in separate silos. Economists proved properties of estimators on whiteboards; data scientists built accurate predictive models that could not answer the question every decision-maker asks: what happens if we intervene?

This course bridges the two. You will use a modern data stack (Polars, DuckDB) and zero-setup cloud notebooks to simulate how data are generated, state your causal assumptions as graphs, estimate effects with experimental, quasi-experimental and machine-learning methods, test how fragile they are, and finally serve a causal model to a dashboard that a policy-maker can use.

The approach is code over calculus: an introductory grasp of probability and regression is enough, because every concept is demonstrated by a simulation whose true answer you know.