Improved Nested Sampling and Surrogate‐Enabled Comparison With Other Marginal Likelihood Estimators
Open Access
- 3 February 2018
- journal article
- research article
- Published by American Geophysical Union (AGU) in Water Resources Research
- Vol. 54 (2) , 797-826
Abstract
Estimating marginal likelihood is of central importance to Bayesian model selection and/or model averaging. The nested sampling method has been recently used together with the Metropolis‐Hasting (M‐H) sampling algorithm for estimating marginal likelihood. This study develops a new implementation of nested sampling by using the DiffeRential Evolution Adaptive Metropolis (DREAMzs) sampling algorithm. The two implementations of nested sampling are evaluated for four models of a synthetic groundwater flow modeling. The DREAMzs‐based nested sampling outperforms the M‐H‐based nested sampling in terms of their accuracy, convergence, efficiency, and stability, which is attributed to the fact that DREAMzs is more robust than M‐H for parameter sampling. The nested sampling method is also compared with four other methods (arithmetic mean, harmonic mean, stabilized harmonic mean, and thermodynamic integration) commonly used for estimating marginal likelihood and posterior probability of the four groundwater models. The comparative study requires hundreds of millions of model executions, which would not be possible without using surrogate models to replace the original models. Using the arithmetic mean estimator as the reference, the comparison reveals that thermodynamic integration outperforms nested sampling in terms of accuracy and stability, whereas nested sampling is more computationally efficient to reach to convergence. The harmonic mean and stabilized harmonic mean methods give the worst marginal likelihood estimation. Accurate marginal likelihood estimation is important for accurate estimation of posterior model probability and better predictive performance of Bayesian model averaging.Keywords
Funding Information
- National Natural Science Foundation of China (U1503282, 41302181, 41672233, 41571017, 51409161)
- Program for New Century Excellent Talents in University (NCET‐12‐0262)
- U.S. Department of Energy (DE‐SC0008272)
- National Science Foundation (1552329)
This publication has 70 references indexed in Scilit:
- Computing the Bayes Factor from a Markov Chain Monte Carlo Simulation of the Posterior DistributionBayesian Analysis, 2012
- Bayesian calibration of a large‐scale geothermal reservoir model by a new adaptive delayed acceptance Metropolis Hastings algorithmWater Resources Research, 2011
- Unified Framework to Evaluate Panmixia and Migration Direction Among Multiple Sampling LocationsGenetics, 2010
- Comment on “Inverse groundwater modeling for hydraulic conductivity estimation using Bayesian model averaging and variance window” by Frank T.‐C. Tsai and Xiaobao LiWater Resources Research, 2010
- Accelerating Markov Chain Monte Carlo Simulation by Differential Evolution with Self-Adaptive Randomized Subspace SamplingInternational Journal of Nonlinear Sciences and Numerical Simulation, 2009
- Nested sampling for general Bayesian computationBayesian Analysis, 2006
- Maximum likelihood Bayesian averaging of uncertain model predictionsStochastic Environmental Research and Risk Assessment, 2003
- A Shuffled Complex Evolution Metropolis algorithm for optimization and uncertainty assessment of hydrologic model parametersWater Resources Research, 2003
- Bayes FactorsJournal of the American Statistical Association, 1995
- Estimating the Dimension of a ModelThe Annals of Statistics, 1978