Reliability, bias, and computational cost of estimating the Bayes factor using bridge sampling and the Savage–Dickey density ratio

Klaus Oberauer, Philipp Musfeld, Frederik Aust · 2026

Read Reliability, bias, and computational cost of estimating the Bayes factor using bridge sampling and the Savage–Dickey density ratio by Klaus Oberauer, Philipp Musfeld, Frederik Aust (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13287297.txt.

Description

Klaus Oberauer, Philipp Musfeld, and Frederik Aust ran six simulation studies in 2026 comparing two ways to estimate a Bayes factor, a measure used to weigh one statistical model against another. One route bridges between models. The other reads a density ratio from a single fitted model.

With many simulation draws, both routes gave accurate answers across wide conditions. Bridging grew slow and shaky as models gained parameters. The density ratio stayed fast and steady at larger sizes, though careless density smoothing could bias it. The authors give practical guidance on matching method to model size and on checking density estimates before trusting the result.

Catalog record

CREATORKlaus Oberauer, Philipp Musfeld, Frederik Aust
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE116793
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13287297

Browse the full library · About the Tower of Babel