Prior-Posterior Derived-Predictive Consistency Checks for Post-Estimation Calculated Quantities of Interest ( QOI-Check )
Holger Sennhenn-Reulen · 2024
Read Prior-Posterior Derived-Predictive Consistency Checks for Post-Estimation Calculated Quantities of Interest ( QOI-Check ) by Holger Sennhenn-Reulen (2024) 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/arxiv-2412-15809.txt.
Description
Holger Sennhenn-Reulen introduces a check for quantities computed after a Bayesian model is fit, in a 2024 paper. Flexible tools such as Stan make such after-fit quantities popular. Examples include group averages through curved links and splits of explained variation. Errors in these extra calculations can mislead users quietly.
His check joins two older validation ideas. One verifies the computing itself by comparing prior and posterior on simulated data. The other tests predictions against held-out real data. Combined, they test both the calculation and its interpretation for a stated population. The goal is a systematic part of Bayesian workflow. Better checks should raise trust in after-fit conclusions, the author argues. The method is presented as a general routine, not tied to one software package.
Catalog record
| CREATOR | Holger Sennhenn-Reulen |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-15809