On the Fully Bayesian Treatment of Latent Gaussian Models using Stochastic Simulations

Filippone, M., Zhong, M. and Girolami, M. (2012) On the Fully Bayesian Treatment of Latent Gaussian Models using Stochastic Simulations. Technical Report. School of Computing Science, University of Glasgow.

Full text not currently available from Enlighten.

Publisher's URL: http://www.dcs.gla.ac.uk/publications/paperdetails.cfm?id=9382

Abstract

Latent Gaussian models (LGMs) are extensively used in data analysis given their flexible modeling capabilities and interpretability. The fully Bayesian treatment of LGMs is usually intractable, and therefore it is necessary to resort to approximations. This paper proposes the use of stochastic simulations based on Markov chain Monte Carlo (MCMC) methods for small to moderately sized data sets and for LGMs comprising a set of parameters that prevents the use of quadrature techniques. We discuss the challenges in applying MCMC methods to LGMs and compare different strategies based on efficient parametrizations and efficient proposal mechanisms. Extensive evaluation on simulated and real data suggests a sampling strategy that achieves high efficiency with moderate cost compared to state-of-the-art methods for the fully Bayesian treatment of LGMs.

Item Type:Research Reports or Papers (Technical Report)
Additional Information:DCS Tech Report Number: TR-2012-329 Paper ID: 9382 Appeared in: DCS Technical Report Series
Status:Published
Glasgow Author(s) Enlighten ID:Zhong, Dr Mingjun and Filippone, Dr Maurizio and Girolami, Prof Mark
Authors: Filippone, M., Zhong, M., and Girolami, M.
College/School:College of Science and Engineering > School of Computing Science
Publisher:School of Computing Science, University of Glasgow

University Staff: Request a correction | Enlighten Editors: Update this record