|
|||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||
SUMMARY: INNER | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
A Metropolis Markov Chain sampler.
Metropolis samplers are Markov Chains that generate the next state by making generating a prospective next state from a known distribution, and then deciding whether to use prospective state by computing the Metropolis acceptance function. This function depends only on the ratio of the target distribution's probability density at the current and proposed points. Hence normalization constants are not required.
the Metropolis sampler is only valid for proposal distributions are symmetric in that P(X|Y) = P(Y|X).
Methods inherited from interface org.omegahat.Simulation.MCMC.MarkovChain |
getState, iterate, step |
|
|||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||
SUMMARY: INNER | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |