New answers tagged gibbs
2
votes
Mathematical Reference for Metropolis-Within-Gibbs Algorithm
Under the positivity assumption required by the Hammersley-Clifford Theorem, the regular Gibbs sampler is stationary wrt the joint distribution and converging.
Adding a Metropolis step instead of ...
3
votes
Mathematical Reference for Metropolis-Within-Gibbs Algorithm
If you take the selected index for updating $i$ to be random with some distribution, there is nothing special about metropolis-hastings-within-gibbs (MHWG). It's quite simply just another MH proposal. ...
Top 50 recent answers are included
Related Tags
gibbs × 419markov-chain-montecarlo × 191
bayesian × 173
sampling × 93
metropolis-hastings × 65
conditional-probability × 41
simulation × 24
posterior × 24
monte-carlo × 23
hierarchical-bayesian × 23
r × 21
markov-process × 20
self-study × 18
machine-learning × 15
probability × 15
normal-distribution × 15
gaussian-mixture-distribution × 13
dirichlet-process × 13
convergence × 12
prior × 12
topic-models × 10
mixture-distribution × 9
jags × 9
dirichlet-distribution × 9
references × 8