Monte Carlo Methods and Applications

Gibbs samplers are common Markov chain Monte Carlo (MCMC) algorithms that used to sample from intractable probability distributions when sampling directly full conditional is possible. These types of MCMC come up frequently in many applications, and because their popularity it important have a sense how long takes for the sampler become close its stationary distribution. To this end, rely on values drift minorization coefficients bound mixing time sampler. This manuscript provides computational method estimating these coefficients. Herein, we detail several advantages proposed methods, as well limitations approach. primarily related “curse dimensionality”, which methods caused by necessary increases numbers initial states chains need be run an exponentially increasing number grid points estimation