{"id":"arxiv/2609.23350","url":"https://arxiv.org/abs/2609.23350","tags":["math.ST","stat.ME"],"score":0,"title":"Retrieving predictive densities from conformal predictive distributions","domain":"math.ST","source":"arxiv","authors":["Dan Andrei Tudor","Dorota Kurowicka","Balint Negyesi","Valerii Zoller"],"reviews":0,"subject":"math.ST","summary":"Conformal prediction intervals are used to construct marginally well-calibrated regions around point predictions. Recent work has focused on extending these ideas to conformal predictive distributions, whose marginal probability integral transform (PIT) follows a Unif[0, 1] distribution. In this work, we investigate how to recover predictive densities from conformal predictive distributions. We extend the theory of conformal predictive distributions by proving asymptotic marginal validity for a tail-corrected version of conformal predictive distributions. We propose a method called quantile…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-09-20","reactions":{}}