Gibbs sampling

Gibbs sampling is a method for sampling from a multivariate target density

Gibbs sampling is a special case of this multivariate generalisation in which the candidate value is always accepted

The Gibbs sampling algorithm has been named by modern-day statisticians after the American physicist Josiah Willard Gibbs (1839–1903), in reference to an analogy between the sampling algorithm and methods of statistical physics.