Bayesian Statistics: Computation
Mathematics Subject Classification
Markov chain Monte Carlo (MCMC)
Markov chain Monte Carlo (MCMC) is a collection of computational methods for simulating from posterior distributions.
Markov chain Monte Carlo (MCMC) methods are a collection of computational algorithms designed to sample from a target distribution by performing Monte Carlo simulation from a Markov chain whose equilibrium distribution is equal to the target distribution. The output of the algorithm is then used to estimate features of the required distribution, where the quality of the estimate is determined by the number of iterations of the algorithm. Surprisingly, it took several decades before the statistical community embraced Markov chain Monte Carlo (MCMC) as a general computational tool in Bayesian inference, where it may be quite difficult to compute the normalizing constant of the (possibly high-dimensional) posterior distribution required for routine...
- 1.Binder, K.: Monte Carlo Methods in Statistical Physics. Springer, New York (1978)Google Scholar
- 9.Morris, C.N.: Comment on “The calculation of posterior distributions by data augmentation” by M.A. Tanner and W.H. Wong. J. Am. Stat. Assoc. 82, 542–543 (1987)Google Scholar
- 10.O’Hagan, A.: Monte Carlo is fundamentally unsound. J. R. Stat. Soc. D: Stat. 36, 247–249 (1987)Google Scholar
- 13.Tanner, M.A.: Metropolis algorithms. In: The Encyclopedia of Applied and Computational Mathematics (2014)Google Scholar