Don’t Disregard the Data for Lack of a Likelihood: Bayesian Synthetic Likelihood for Enhanced Multilevel Network Meta-Regression
2026
Read Don’t Disregard the Data for Lack of a Likelihood: Bayesian Synthetic Likelihood for Enhanced Multilevel Network Meta-Regression (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/arxiv-2603-11019.txt.
Description
Harlan Campbell and coauthors extend multilevel network meta-regression so it can use subgroup results that standard versions throw away, in a method for population-adjusted indirect comparisons of treatments. Their Bayesian approach rests on a synthetic likelihood. It fills in missing covariates at each sampling step and builds synthetic subgroup summaries. It matches them to the real ones through a normal synthetic likelihood. Pre-drawn random numbers and continuous relaxation let the scheme run in Stan's Hamiltonian Monte Carlo. A psoriasis trial network shows gains over the standard method.
Catalog record
| YEAR | 2026 |
|---|---|
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64911 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.11019