Paul R. Miles
Department of Mathematics, North Carolina State University, Raleigh, NC 27695

Ralph C. Smith
Department of Mathematics, North Carolina State University, Raleigh, NC 27695

Abstract

A Bayesian approach to solving inverse problems provides insight regarding model limitations as well as the underlying model and observation uncertainty. In this paper we introduce pymcmcstat, which provides a wide variety of tools for estimating unknown parameter distributions. For scientists and engineers familiar with least-squares optimization, this package provides a similar interface from which to expand their analysis to a Bayesian framework. This package has been utilized in a wide array of scientific and engineering problems, including radiation source localization and constitutive model development of smart material systems.

Keywords

Markov Chain Monte Carlo (MCMC), Delayed Rejection Adaptive Metropolis (DRAM), Parameter Estimation, Bayesian Inference

DOI

10.25080/Majora-7ddc1dd1-00d

Bibtex entry

Full text PDF

Copyright © 2019 Paul R. Miles et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.