Direct comparison between Bayesian and frequentist uncertainty quantification for nuclear reactions

Abstract

Until recently, uncertainty quantification in low energy nuclear theory was typically performed using frequentist approaches. However in the last few years, the field has shifted toward Bayesian statistics for evaluating confidence intervals. Although there are statistical arguments to prefer the Bayesian approach, no direct comparison is available. In this work, we compare, directly and systematically, the frequentist and Bayesian approaches to quantifying uncertainties in direct nuclear reactions. Starting from identical initial assumptions, we determine confidence intervals associated with the elastic and the transfer process for both methods, which are evaluated against data via a comparison of the empirical coverage probabilities. Expectedly, the frequentist approach is not as flexible as the Bayesian approach in exploring parameter space and often ends up in a different minimum. We also show that the two methods produce significantly different correlations. In the end, the frequentist approach produces significantly narrower uncertainties on the considered observables than the Bayesian. Our study demonstrates that the uncertainties on the reaction observables considered here within the Bayesian approach represent reality more accurately than the much narrower uncertainties obtained using the standard frequentist approach.

accepted for publication in Phys. Rev. Letts

References in corpus (8)

• Quantifying truncation errors in effective field theory
• Bayesian approach to model-based extrapolation of nuclear observables
• Neutron star tidal deformabilities constrained by nuclear theory and experiment
• Bayesian truncation errors in chiral effective field theory: nucleon-nucleon observables
• Constraining Transfer Cross Sections Using Bayes' Theorem
• Uncertainty Quantification for Optical Model Parameters
• Transfer reaction code with nonlocal interactions
• Uncertainty quantification due to optical potentials in models for (d,p) reactions

Cited by in corpus (33)

• Machine Learning in Nuclear Physics
• Get on the BAND Wagon: A Bayesian Framework for Quantifying Model Uncertainties in Nuclear Dynamics
• Nuclear masses learned from a probabilistic neural network
• Optical potentials for the rare-isotope beam era
• Toward emulating nuclear reactions using eigenvector continuation
• Model reduction methods for nuclear emulators
• Nuclear Forces for Precision Nuclear Physics -- a collection of perspectives
• Performing Bayesian analyses with AZURE2 using BRICK: an application to the ${}^7$Be system
• A global microscopic description of nucleon-nucleus scattering with quantified uncertainties
• Uncertainty-quantified phenomenological optical potentials for single-nucleon scattering
• Fast emulation of quantum three-body scattering
• Recent advances in the quantification of uncertainties in reaction theory
• Ab initio prediction of the ${}^4\mathrm{He}(d,\gamma ){}^6\mathrm{Li}$ big bang radiative capture
• Isotopically resolved neutron total cross sections at intermediate energies
• Exploring experimental conditions to reduce uncertainties in the optical potential
• New Test of Modulated Electron Capture Decay of Hydrogen-Like ${}^{142}$Pm Ions: Precision Measurement of Purely Exponential Decay
• Solar fusion III: New data and theory for hydrogen-burning stars
• Halo effective field theory analysis of one-neutron knockout reactions of ${}^{11}\mathrm{Be}$ and ${}^{15}\mathrm{C}$
• Statistical tools for a better optical model
• Mean shift cluster recognition method implementation in the nested sampling algorithm
• Prediction of (p,n) Charge-Exchange Reactions with Uncertainty Quantification
• Microscopic optical potential from the relativistic Brueckner-Hartree-Fock theory: Proton-nucleus scattering
• The Nested_fit data analysis program
• Quantifying uncertainties due to optical potentials in one-neutron knockout reactions
• The complete quantification of parametric uncertainties in (d,p) transfer reactions
• A Complex Scaling Method for Efficient and Accurate Scattering Emulation in Nuclear Reactions
• A Fast and Calibrated Computer Model Emulator: An Empirical Bayes Approach
• Bayesian Analysis of the ${}^{70}$Zn$(d{,}^3\text{He}{)}^{69}$Cu Transfer Reaction
• Report from the A.I. For Nuclear Physics Workshop
• First application of Markov Chain Monte Carlo-based Bayesian data analysis to the Doppler-Shift Attenuation Method
• Study of Cluster Structures in Nuclei through the Ratio Method. A Tribute to Mahir Hussein
• Uncertainty Quantification in Breakup Reactions
• Gaussian process error modeling for chiral effective-field-theory calculations of $np\leftrightarrow d\gamma$ at low energies