Model averaging approaches to data subset selection
arXiv:2305.19417 · doi:10.1103/PhysRevE.108.045308
Abstract
Model averaging is a useful and robust method for dealing with model uncertainty in statistical analysis. Often, it is useful to consider data subset selection at the same time, in which model selection criteria are used to compare models across different subsets of the data. Two different criteria have been proposed in the literature for how the data subsets should be weighted. We compare the two criteria closely in a unified treatment based on the Kullback-Leibler divergence, and conclude that one of them is subtly flawed and will tend to yield larger uncertainties due to loss of information. Analytical and numerical examples are provided.
arXiv comment17 pages, 2 figures. v2: updated to published version
References in corpus (6)
- A high-bias, low-variance introduction to Machine Learning for physicists
- Leading hadronic contribution to the muon magnetic moment from lattice QCD
- Ab initio calculation of the neutron-proton mass difference
- Constrained Curve Fitting
- -mixing matrix elements from lattice QCD for the Standard Model and beyond
- Bayesian model averaging for analysis of lattice field theory results
Cited by in corpus (11)
- Improved information criteria for Bayesian model averaging in lattice field theory
- Nucleon Sigma Terms with O()-improved Wilson fermions
- form factors from lattice QCD with domain-wall quarks: A new piece in the puzzle of decay rates
- Physical-mass calculation of and resonance parameters via and scattering amplitudes from lattice QCD
- Deep learning of phase transitions with minimal examples
- A generalized statistical model for fits to parton distributions
- The scattering amplitude at large
- Single Particle Spectrum of Doped -Perylene
- form factors from lattice QCD and Standard-Model predictions for and decays
- The two-pion contribution to the hadronic vacuum polarization with staggered quarks
- in iso-symmetric QCD and the CKM matrix unitarity