Doubly robust and efficient calibration of prediction sets for right-censored time-to-event outcomes
Rebecca Farina, Eric Tchetgen Tchetgen, Arun Kumar Kuchibhotla · 2026
Read Doubly robust and efficient calibration of prediction sets for right-censored time-to-event outcomes by Rebecca Farina, Eric Tchetgen Tchetgen, Arun Kumar Kuchibhotla (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/PMC13433195.txt.
Description
Rebecca Farina, Eric Tchetgen Tchetgen, and Arun Kumar Kuchibhotla built math tools in 2026 to set honest forecast bands for survival times when some study exits hide the true end. A person's event time may stay unseen when they quit a study or the study ends first. Past tools for such bands leaned on a narrow kind of hidden exit with full exit times known.
Their route needs no firm shape for the survival curve. Several low bands rest on weights that fix for hidden exits, and one boosted band draws on a smart score built for the target mark. Proofs show the band holds its promised cover rate in the long run. The boosted band's cover slip shrinks at a faster order than its kin, which makes its error fade fast as data grows. Tests back the edge.
Catalog record
| CREATOR | Rebecca Farina, Eric Tchetgen Tchetgen, Arun Kumar Kuchibhotla |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 142822 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |