Bayesian analysis of ODEs: solver optimal accuracy and Bayes factors
From MaRDI portal
Publication:3179310
DOI10.1137/140976777zbMATH Open1352.65193arXiv1311.2281OpenAlexW310515683MaRDI QIDQ3179310FDOQ3179310
Authors: Marcos A. Capistrán, J. A. Christen, Sophie Donnet 
Publication date: 21 December 2016
Published in: SIAM/ASA Journal on Uncertainty Quantification (Search for Journal in Brave)
Abstract: In most relevant cases in the Bayesian analysis of ODE inverse problems, a numerical solver needs to be used. Therefore, we cannot work with the exact theoretical posterior distribution but only with an approximate posterior deriving from the error in the numerical solver. To compare a numerical and the theoretical posterior distributions we propose to use Bayes Factors (BF), considering both of them as models for the data at hand. We prove that the theoretical vs a numerical posterior BF tends to 1, in the same order (of the step size used) as the numerical forward map solver does. For higher order solvers (eg. Runge-Kutta) the Bayes Factor is already nearly 1 for step sizes that would take far less computational effort. Considerable CPU time may be saved by using coarser solvers that nevertheless produce practically error free posteriors. Two examples are presented where nearly 90% CPU time is saved while all inference results are identical to using a solver with a much finer time step.
Full work available at URL: https://arxiv.org/abs/1311.2281
Recommendations
- Efficient Bayesian estimation and uncertainty quantification in ordinary differential equation models
- Bayesian ODE solvers: the maximum a posteriori estimate
- scientific article; zbMATH DE number 802686
- Bayesian inference for higher-order ordinary differential equation models
- Adaptive step-size selection for state-space probabilistic differential equation solvers
Bayesian inference (62F15) Applications of statistics (62P99) Multistep, Runge-Kutta and extrapolation methods for ordinary differential equations (65L06) Numerical solution of inverse problems involving ordinary differential equations (65L09)
Cites Work
- A general purpose sampling algorithm for continuous distributions (the t-walk)
- Bayesian model averaging: A tutorial. (with comments and a rejoinder).
- Title not available (Why is that?)
- An adaptive Metropolis algorithm
- Statistical and computational inverse problems.
- Bayes Factors
- Title not available (Why is that?)
- Sieve estimation of constant and time-varying coefficients in nonlinear ordinary differential equation models by considering both numerical error and measurement error
- Title not available (Why is that?)
- On adaptive Markov chain Monte Carlo algorithms
- Numerical mathematics.
- A First Course in the Numerical Analysis of Differential Equations
- Title not available (Why is that?)
- Linear multistep methods, particle filtering and sequential Monte Carlo
- A Posteriori Error Estimation and Global Error Control for Ordinary Differential Equations by the Adjoint Method
- An Introduction to Numerical Analysis
- Approximation of Bayesian Inverse Problems for PDEs
- A Bayesian approach to multiscale inverse problems using the sequential Monte Carlo method
- Sparse deterministic approximation of Bayesian inverse problems
- RECONSTRUCTION OF DOMAIN BOUNDARY AND CONDUCTIVITY IN ELECTRICAL IMPEDANCE TOMOGRAPHY USING THE APPROXIMATION ERROR APPROACH
- Estimation of parameters in incomplete data models defined by dynamical systems
- Non-stationary magnetoencephalography by Bayesian filtering of dipole models
- On Global Error Estimation and Control for Initial Value Problems
- Parameter estimation for stiff deterministic dynamical systems via ensemble Kalman filter
- Title not available (Why is that?)
- Local regularization method applied to estimating oxygen consumption during muscle activities
- Bayesian MLP neural networks for image analysis
- Towards uncertainty quantification and inference in the stochastic SIR epidemic model
Cited In (11)
- Statistical analysis of differential equations: introducing probability measures on numerical solutions
- Bayes linear analysis for ordinary differential equations
- Penalised t-walk MCMC
- Bayesian Mesh Adaptation for Estimating Distributed Parameters
- Bayesian Probabilistic Numerical Methods
- Strong convergence rates of probabilistic integrators for ordinary differential equations
- Disentangling the role of virus infectiousness and awareness-based human behavior during the early phase of the COVID-19 pandemic in the European Union
- Bayesian analysis for a fractional population growth model
- Bayesian solution uncertainty quantification for differential equations
- Error control of the numerical posterior with Bayes factors in Bayesian uncertainty quantification
- An importance sampling approach for reliable and efficient inference in Bayesian ordinary differential equation models
Uses Software
This page was built for publication: Bayesian analysis of ODEs: solver optimal accuracy and Bayes factors
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3179310)
