Author： Bliznyuk Nikolay Ruppert David Shoemaker Christine A.
Publisher： Taylor & Francis Ltd
ISSN： 1061-8600
Source： Journal of Computational and Graphical Statistics, Vol.20, Iss.3, 2011-01, pp. : 636-655
Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.
Abstract
Markov chain Monte Carlo (MCMC) is nowadays a standard approach to numerical computation of integrals of the posterior density π of the parameter vector η. Unfortunately, Bayesian inference using MCMC is computationally intractable when the posterior density π is expensive to evaluate. In many such problems, it is possible to identify a minimal subvector β of η responsible for the expensive computation in the evaluation of π. We propose two approaches, DOSKA and INDA, that approximate π by interpolation in ways that exploit this computational structure to mitigate the curse of dimensionality. DOSKA interpolates π directly while INDA interpolates π indirectly by interpolating functions, for example, a regression function, upon which π depends. Our primary contribution is derivation of a Gaussian processes interpolant that provably improves over some of the existing approaches by reducing the effective dimension of the interpolation problem from dim(η) to dim(β). This allows a dramatic reduction of the number of expensive evaluations necessary to construct an accurate approximation of π when dim(η) is high but dim(β) is low. We illustrate the proposed approaches in a case study for a spatio-temporal linear model for air pollution data in the greater Boston area. Supplemental materials include proofs, details, and software implementation of the proposed procedures.
Related content
Access to resources
Recommend
By Hesthaven Jan S. Stamm Benjamin Zhang Shun
ESAIM: Mathematical Modelling and Numerical Analysis, Vol. 48, Iss. 1, 2014-01 ,pp. :