$ ⌘K
// constructs / markov_chain_monte_carlo

Markov Chain Monte Carlo (MCMC)

id: markov_chain_monte_carlo

A class of algorithms (Metropolis-Hastings, Gibbs sampling) that generate samples from an arbitrary target distribution — typically a Bayesian posterior — by constructing a Markov chain whose stationary distribution is the target. The dominant approach to approximate posterior inference when the posterior lacks a closed form.

// usage

Used in 1 pax

↑ positive relationship 2 findings