Modern Model-Based Bayesian Causal Inference for Randomized Experiments with Hamiltonian Monte Carlo in Stan
Author: Adam Rohde
Date: June 10, 2020
Description: We explore the components of modern model-based Bayesian causal inferenece with a focus on randomized experiments. We discuss the potential outcomes framework, the Bayesian approach to causal inference, the MCMC sampling method Hamiltonian Monte Carlo, and the Stan probabilistic programming language. We also work through a simple example to illustrate how these components come together.
Files:
- Report: Bayesian-Causal-Inference-for-Randomized-Experiments-with-HMC.pdf
- R Code: Bayesian-Causal-Inference-for-Randomized-Experiments-with-HMC.Rmd