Posterior Approximation by Interpolation for Bayesian Inference in Computationally Expensive Statistical Models

Other Titles


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 π is expensive to evaluate. In this work, we develop practical methods that approximate π with radial basis functions (RBFs) and Gaussian processes (GPs) interpolants and use the resulting cheap-to-evaluate surfaces in MCMC.

In Chapter 1, π arises from a nonlinear regression model with transformation and dependence. To build the RBF approximation, we limit evaluation of the computationally expensive regression function to points chosen on a high posterior density (HPD) region found using a local quadratic approximation of log⁡(π) at its mode. We illustrate our approach on simulated data for a pollutant diffusion problem and study the frequentist coverage properties of credible intervals.

In Chapter 2, we relax the assumptions about π made in Chapter 1 and develop a derivative-free procedure GRIMA to approximate the logarithm of π using RBF interpolation over a HPD region of π estimated using the RBF surface. We use GRIMA for Bayesian inference in a computationally intensive nonlinear regression model for real measured streamflow data in the Town Brook watershed.

In Chapter 3, we study statistical models where 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 to approximate π by interpolation that exploit this computational structure. Our primary contribution is derivation of a GP 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). When dim⁡(η) is high but dim⁡(β) is low, this allows one to dramatically reduce the number of expensive evaluations necessary to construct an accurate approximation of π.

Our experiments indicate that our methods produce results similar to those when the true expensive posterior density is sampled by MCMC while reducing computational costs by well over an order of magnitude.

Journal / Series

Volume & Issue



NSF DMS-04-538

Date Issued




inverse problems; response surface


Effective Date

Expiration Date




Union Local


Number of Workers

Committee Chair

Committee Co-Chair

Committee Member

Degree Discipline

Degree Name

Degree Level

Related Version

Related DOI

Related To

Related Part

Based on Related Item

Has Other Format(s)

bibid: 6397183

Part of Related Item

Related To

Related Publication(s)

Link(s) to Related Publication(s)


Link(s) to Reference(s)

Previously Published As

Government Document




Other Identifiers


Rights URI


dissertation or thesis

Accessibility Feature

Accessibility Hazard

Accessibility Summary

Link(s) to Catalog Record