Category:Definitions/Gibbs Sampler

This category contains definitions related to Gibbs Sampler.
Related results can be found in Category:Gibbs Sampler.

The Gibbs sampler is a Markov chain Monte Carlo technique used to give numerical approximations to Bayesian posterior distributions involving $2$ or more variables.

An initial set of values is specified, and new values of each variable are successively simulated from their conditional distributions, given the current values of all other variables.

If the new value is more in accord with the specified distribution, it replaces the current value, otherwise the current value is kept.

The process is continued until an equilibrium is reached.