Gibbs sampling

In statistics, Gibbs sampling or a Gibbs sampler is a Markov chain Monte Carlo (MCMC) algorithm for obtaining a sequence of observations which are approximated from a specified multivariate probability distribution, when direct sampling is difficult. This sequence can be used to approximate the joint distribution (e.g., to generate a histogram of the distribution); to approximate the marginal distribution of one of the variables, or some subset of the variables (for example, the unknown parameters or latent variables); or to compute an integral (such as the expected value of one of the variables). Typically, some of the variables correspond to observations whose values are known, and hence do not need to be sampled.

Comment: enIn statistics, Gibbs sampling or a Gibbs sampler is a Markov chain Monte Carlo (MCMC) algorithm for obtaining a sequence of observations which are approximated from a specified multivariate probability distribution, when direct sampling is difficult. This sequence can be used to approximate the joint distribution (e.g., to generate a histogram of the distribution); to approximate the marginal distribution of one of the variables, or some subset of the variables (for example, the unknown parameters or latent variables); or to compute an integral (such as the expected value of one of the variables). Typically, some of the variables correspond to observations whose values are known, and hence do not need to be sampled.
Depiction
Has abstract: enIn statistics, Gibbs sampling or a Gibbs sampler is a Markov chain Monte Carlo (MCMC) algorithm for obtaining a sequence of observations which are approximated from a specified multivariate probability distribution, when direct sampling is difficult. This sequence can be used to approximate the joint distribution (e.g., to generate a histogram of the distribution); to approximate the marginal distribution of one of the variables, or some subset of the variables (for example, the unknown parameters or latent variables); or to compute an integral (such as the expected value of one of the variables). Typically, some of the variables correspond to observations whose values are known, and hence do not need to be sampled. Gibbs sampling is commonly used as a means of statistical inference, especially Bayesian inference. It is a randomized algorithm (i.e. an algorithm that makes use of random numbers), and is an alternative to deterministic algorithms for statistical inference such as the expectation-maximization algorithm (EM). As with other MCMC algorithms, Gibbs sampling generates a Markov chain of samples, each of which is correlated with nearby samples. As a result, care must be taken if independent samples are desired. Generally, samples from the beginning of the chain (the burn-in period) may not accurately represent the desired distribution and are usually discarded.
Hypernym: Algorithm
Is primary topic of: Gibbs sampling
Label: enGibbs sampling
Link from a Wikipage to an external page: turing.ml/
Link from a Wikipage to another Wikipage: Adaptive rejection sampling; Algorithm; American Mathematical Society; Andrew Gelman; Autocorrelation; Bayes estimator; Bayesian analysis; Bayesian inference; Bayesian learning; Bayesian network; Bayesian statistics; Black magic (programming); Cartesian product; Categorical distribution; Category:Markov chain Monte Carlo; Chapman & Hall; Church (programming language); Compound distribution; Conditional distribution; Conjugate prior; Continuous distribution; Curse of dimensionality; Detailed balance equations; Deterministic algorithm; Deterministically; Dirichlet distribution; Dirichlet-multinomial distribution; Discrete distribution; Donald Geman; Elizabeth Wilmer; Equivalence relation; Expectation maximization; Expectation-maximization; Expectation-maximization algorithm; Expected value; Exponential family; File:Gibbs sampler picture.jpg; File:Gibbs sampling info eq.jpg; Forward-backward algorithm; Gamma distribution; Gaussian distribution; Generalized linear model; Generalized linear models; Graphical model; Hidden Markov model; Hierarchical Bayesian model; Independent identically distributed; Integral; Inverse gamma distribution; Joint distribution; Josiah Willard Gibbs; Journal of the American Statistical Association; Julia (programming language); Just another Gibbs sampler; Latent Dirichlet allocation; Latent variable; Linear regression; Logarithmically concave function; Logistic function; Logistic regression; Marginal distribution; Markov chain; Markov chain Monte Carlo; Markov Chains and Mixing Times; Maximum a posteriori; Maximum entropy classifier; Mean; Metropolis–Hastings; Metropolis–Hastings algorithm; Mixture model; Mode (statistics); Multivariate distribution; Natural language processing; Negative binomial distribution; Normal distribution; Normalization constant; OpenBUGS; Parameter; Poisson distribution; Posterior distribution; Posterior predictive distribution; Posterior probability; Prior distribution; Probabilistic programming; Probability distribution; Probit regression; PyMC; Python (programming language); Randomized algorithm; Random number generation; Random variables; Random walk; Sample mean; Sample variance; Sampling (statistics); Semi-supervised learning; Simulated annealing; Skewness; Slice sampling; Stationary distribution; Statistical inference; Statistical physics; Statistics; Stuart Geman; Student's t-distribution; Supervised learning; Topic model; Unsupervised learning; Variance; Variational Bayes; Yuval Peres
SameAs: Amostragem de Gibbs; Campionamento di Gibbs; Échantillonnage de Gibbs; EZBC; Gibbs sampling; Gibbs-Sampling; m.02jxh7; Muestreo de Gibbs; Q1191905; Семплирование по Гиббсу; نمونه‌برداری گیبز; ギブスサンプリング; 吉布斯采样; 기브스 표집
Subject: Category:Markov chain Monte Carlo
Thumbnail
WasDerivedFrom: Gibbs sampling?oldid=1118061637&ns=0
WikiPageLength: 37415
Wikipage page ID: 509709
Wikipage revision ID: 1118061637
WikiPageUsesTemplate: Template:Bayesian statistics; Template:Citation; Template:Cite journal; Template:ISBN; Template:Reflist; Template:Short description

Gibbs sampling

Backlinks

About

Resources

Support

Follow us