Parameter Exploration in Simulation Experiments: A Bayesian Framework

TL;DR

This paper introduces a Bayesian Markov chain Monte Carlo framework for exploring uncertain parameters in simulation experiments, addressing limitations of grid-based methods and improving parameter space coverage.

Contribution

It proposes a novel MCMC-based approach for parameter exploration in simulations, enhancing the ability to identify important parameter combinations beyond fixed grids.

Findings

Effective exploration of parameter space demonstrated in phylogenetics, archaeology, and epidemiology examples.

Addresses limitations of grid-based parameter selection methods.

Provides a flexible, probabilistic framework for simulation parameter analysis.

Abstract

Simulations often involve the use of model parameters which are unknown or uncertain. For this reason, simulation experiments are often repeated for multiple combinations of parameter values, often iterating through parameter values lying on a fixed grid. However, the use of a discrete grid places limits on the dimension of the parameter space and creates the potential to miss important parameter combinations which fall in the gaps between grid points. Here we draw parallels with strategies for numerical integration and describe a Markov chain Monte-Carlo strategy for exploring parameter values. We illustrate the approach using examples from phylogenetics, archaeology, and epidemiology.