This textbook provides a practical guide to the Bayesian framework for data modeling and causal inference, focusing on model interpretation, diagnostics, and uncertainty quantification. Central to the book is a "learning-by-doing" approach, using concrete examples in R with real-world datasets spanning diverse fields, including education, psychology, medicine, behavioral science, and environmental science. The book is structured into three parts:· Part I: Linear Regression – Learn the basics of Bayesian linear regression, model diagnostics, and uncertainty quantification through a probabilistic lens. · Part II: Generalized Linear Models – Extend your modeling toolkit to handle binary and count data, zero-inflated models, and clustered data structures common in longitudinal studies. · Part III: Causal Inference – Learn to identify treatment effects from non-experimental data. This section explores classical techniques—including inverse probability weighting, doubly robust estimation, instrumental variables, and difference-in-differences—alongside advanced techniques like synthetic control, doubly robust DiD, and synthetic DiD.
