Time-to-Event Modeling with Pseudo-Observations in Federated Settings
Hyojung Jang · 2025
Read Time-to-Event Modeling with Pseudo-Observations in Federated Settings by Hyojung Jang (2025) 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-2507-20558.txt.
Description
Hospitals that cannot share patient records can still study survival together if each site shares only safe summaries. Hyojung Jang and co-authors proposed a one-shot federated method for that task in 2025. Their approach builds pseudo-observations from a stepwise survival curve and fits them with an estimating equation, all without pooling raw data or repeated rounds of messages between sites.
The method allows flexible link choices and handles risks that change over time, unlike classic proportional hazards models. A debiasing step shrinks noisy local quirks toward the global estimate while keeping real site differences. Tests on simulations and on pediatric obesity data from a Chicago network of nearly forty-six thousand patients gave robust results.
Catalog record
| CREATOR | Hyojung Jang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-20558