Causal Inference in the Wild

Causal inference is a hot topic in machine learning, and there are many excellent primers on the theory of causal inference available [1–4]. But much fewer examples of real-world applications of machine-learning-powered causal inference exist. This article introduces one such example from an industry context, using a (public) real-world dataset. It is aimed at a technical audience with an understanding of the basics of causality. Specifically, I will look at the "ideal" scenario of price elasticity estimation [2, 5].