Statistics and Computing

, Volume 17, Issue 4, pp 323–335

Parallelizing MCMC for Bayesian spatiotemporal geostatistical models

  • Jun Yan
  • Mary Kathryn Cowles
  • Shaowen Wang
  • Marc P. Armstrong
Article

DOI: 10.1007/s11222-007-9022-2

Cite this article as:
Yan, J., Cowles, M.K., Wang, S. et al. Stat Comput (2007) 17: 323. doi:10.1007/s11222-007-9022-2

Abstract

When MCMC methods for Bayesian spatiotemporal modeling are applied to large geostatistical problems, challenges arise as a consequence of memory requirements, computing costs, and convergence monitoring. This article describes the parallelization of a reparametrized and marginalized posterior sampling (RAMPS) algorithm, which is carefully designed to generate posterior samples efficiently. The algorithm is implemented using the Parallel Linear Algebra Package (PLAPACK). The scalability of the algorithm is investigated via simulation experiments that are implemented using a cluster with 25 processors. The usefulness of the method is illustrated with an application to sulfur dioxide concentration data from the Air Quality System database of the U.S. Environmental Protection Agency.

Keywords

Bayesian inference Markov chain Monte Carlo Parallel computing Spatial modeling 

Copyright information

© Springer Science+Business Media, LLC 2007

Authors and Affiliations

  • Jun Yan
    • 1
  • Mary Kathryn Cowles
    • 1
    • 2
  • Shaowen Wang
    • 3
  • Marc P. Armstrong
    • 4
    • 5
  1. 1.Department of Statistics and Actuarial ScienceThe University of IowaIowaUSA
  2. 2.Department of BiostatisticsThe University of IowaIowaUSA
  3. 3.Department of Geography and National Center for Supercomputing ApplicationsUniversity of Illinois at Urbana-ChampaignIllinoisUSA
  4. 4.Department of GeographyThe University of IowaIowaUSA
  5. 5.Program in Applied Mathematical and Computational ScienceThe University of IowaIowaUSA