Collapsed Gibbs Sampling Algorithm

From GM-RKB
Jump to navigation Jump to search

A Collapsed Gibbs Sampling Algorithm is a Gibbs sampling algorithm that marginalizes over (integrates over) one or more variables when sampling for some other variable.



References

2011

1994

  • (Liu, 1994) ⇒ Jun S. Liu. (1994). “The Collapsed Gibbs Sampler in Bayesian Computations with Applications to a Gene Regulation Problem.” In: Journal of the American Statistical Association, 89(427). http://www.jstor.org/stable/2290921
    • ABSTRACT: This article describes a method of "grouping" and "collapsing" in using the Gibbs sampler and proves from an operator theory viewpoint that the method is in general beneficial. The norms of the forward operators associated with the corresponding non-reversible Markov chains are used to discriminate among different simulation schemes. When applied to Bayesian missing data problems, the idea of collapsing suggests skipping the steps of sampling parameter(s) values in standard data augmentation. By doing this, we obtain a predictive update version of the Gibbs sampler. A procedure of calculating the posterior odds ratio via the collapsed Gibbs sampler when incomplete observations are involved is presented. As an illustration of possible applications, three examples, along with a Bayesian treatment for identifying common protein binding sites in unaligned DNA sequences, are provided.