Cargando…

The Hamming Ball Sampler

We introduce the Hamming ball sampler, a novel Markov chain Monte Carlo algorithm, for efficient inference in statistical models involving high-dimensional discrete state spaces. The sampling scheme uses an auxiliary variable construction that adaptively truncates the model space allowing iterative...

Descripción completa

Detalles Bibliográficos
Autores principales: Titsias, Michalis K., Yau, Christopher
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Taylor & Francis 2017
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5796496/
https://www.ncbi.nlm.nih.gov/pubmed/29456460
http://dx.doi.org/10.1080/01621459.2016.1222288
Descripción
Sumario:We introduce the Hamming ball sampler, a novel Markov chain Monte Carlo algorithm, for efficient inference in statistical models involving high-dimensional discrete state spaces. The sampling scheme uses an auxiliary variable construction that adaptively truncates the model space allowing iterative exploration of the full model space. The approach generalizes conventional Gibbs sampling schemes for discrete spaces and provides an intuitive means for user-controlled balance between statistical efficiency and computational tractability. We illustrate the generic utility of our sampling algorithm through application to a range of statistical models. Supplementary materials for this article are available online.