Computational Bayesian Statistics: Markov Chain Monte Carlo (MCMC) – Practice Quiz

Learn how posterior computation works and how to judge the reliability of its output. Calculate Metropolis-Hastings acceptance probabilities, compare Gibbs and Hamiltonian methods, and interpret chain agreement, bulk and tail ESS, Monte Carlo error, divergences and energy diagnostics. Prerequisites: conditional distributions, posterior densities, gradients and basic Markov-chain ideas. Includes diagnostic tradeoffs and counterexamples to simplistic convergence claims. Single-select and multiselect practice includes worked explanations and distractor rationales.

Study sources: Stan: Posterior Analysis; Stan: MCMC Sampling.

Photo by ThisIsEngineering on Pexels.