10. In Pursuit of the Experiment: Natural Experiments and Difference-in-Difference Modeling
In the last section on predictive methods, we covered both machine learning and demographic forecasting models. These models help us predict user behavior and forecast subpopulation changes and business needs. While predictive insights are useful, they cannot help us change user behavior.
In this section, we’ll examine causal inference methods. Causal inference is an exploration of factors that determine an outcome. Understanding causation will allow us to run better campaigns, build better products, and alter user behavior.
As discussed in Chapter 6 on A/B testing, randomness is the reasoned basis of inference according to Ronald Fisher the father of ...
Get Product Analytics: Applied Data Science Techniques for Actionable Consumer Insights now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
