On the Conservativeness of Robust Variance Estimators in Propensity Score Weighted Cox Models
Hiroya Morita, Shunichiro Orihara, Fumitaka Shimizu, Masataka Taguri · 2026
Read On the Conservativeness of Robust Variance Estimators in Propensity Score Weighted Cox Models by Hiroya Morita, Shunichiro Orihara, Fumitaka Shimizu, Masataka Taguri (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/PMC13621226.txt.
Description
Hiroya Morita, Shunichiro Orihara, Fumitaka Shimizu, and Masataka Taguri examine a shortcut in causal statistics that may not be as safe as researchers assume. Observational studies often weight patients by their chance of receiving a treatment, then fit a survival model. Most software estimates the uncertainty while ignoring that the weights were themselves estimated, a shortcut known to play safe when the target is the average effect across everyone.
The team asked whether that safety survives when the target changes to the effect in treated patients, or in patients where the groups overlap. Math, simulations, and a real dataset all gave the same answer: the uncorrected estimate can run too tight rather than too wide under those other weights. Confidence intervals built on it could claim more certainty than the data hold. The authors recommend variance methods that account for weight estimation whenever such models leave the standard target.
Catalog record
| CREATOR | Hiroya Morita, Shunichiro Orihara, Fumitaka Shimizu, Masataka Taguri |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 13 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |