Frequentist tests for Bayesian models

Abstract

Analogues of the frequentist chi-square and F tests are proposed for testing goodness-of-fit and consistency for Bayesian models. Simple examples exhibit these tests' detection of inconsistency between consecutive experiments with identical parameters, when the first experiment provides the prior for the second. In a related analysis, a quantitative measure is derived for judging the degree of tension between two different experiments with partially overlapping parameter vectors.

8 pages, 4 figures. Section 8 rewritten. Additional references. Accepted by Astronomy & Astrophysics

References in corpus (1)

• Analysing weak orbital signals in Gaia data

Cited by in corpus (11)

• Dust attenuation evolution in $z\sim 2$-$12$ JWST galaxies
• Matter ejections behind the highs and lows of the transitional millisecond pulsar PSR J1023+0038
• Searching for quasi-periodic oscillations in astrophysical transients using Gaussian processes
• Probing jet launching in neutron star X-ray binaries: the variable and polarized jet of SAX J1808.4-3658
• Binary orbits from combined astrometric and spectroscopic data
• Quiet Sun electron densities and their uncertainties derived from spectral emission line intensities
• Bayesian model checking: A comparison of tests
• Bayesian jackknife tests with a small number of subsets: Application to HERA 21cm power spectrum upper limits
• A Multi-wavelength Characterization of the 2023 Outburst of MAXI J1807+132: Manifestations of Disk Instability and Jet Emission
• The most distant optically polarised GRB afterglow: GRB 240419A at z = 5.178
• Resolving dust and Lyα emission in a lensed galaxy at the epoch of reionization with JWST/CANUCS