Prediction Sets and Conformal Inference with Interval Outcomes
Weiguang Liu, Áureo de Paula, Elie Tamer · 2025
Read Prediction Sets and Conformal Inference with Interval Outcomes by Weiguang Liu, Áureo de Paula, Elie Tamer (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-2501-10117.txt.
Description
Weiguang Liu, Áureo de Paula, and Elie Tamer wrote this 2025 statistics paper on prediction sets when the outcome is only known within an interval. Such data arise in job postings with salary ranges and in surveys with bracketed answers. A prediction set should trap a fresh outcome with a stated probability, and the best one is the smallest.
The authors characterize that ideal set under the partial knowledge that intervals create. They give estimators that settle on it as data grows, for both single intervals and unions of disjoint ones. A conformal layer, with a score built for interval data, adds exact finite-sample coverage. The method separates three kinds of uncertainty: chance, partial knowledge, and sampling. Tests use UK job ads and a US population survey.
Catalog record
| CREATOR | Weiguang Liu, Áureo de Paula, Elie Tamer |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 105 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-10117