We consider the conformalization of a stacked ensemble of predictive models, showing that the potentially simple form of the meta-learner at the top of the stack enables a procedure with manageable computational cost that achieves approximate marginal validity without requiring the use of a separate calibration sample. Empirical results indicate that the method compares favorably to a standard inductive alternative.
View on arXiv@article{f2025_2505.12578,
title={ Stacked conformal prediction },
author={ Paulo C. Marques F },
journal={arXiv preprint arXiv:2505.12578},
year={ 2025 }
}