Id: arxiv/2602.09356 Url: https://arxiv.org/abs/2602.09356 Tags: math.ST Score: 0 Title: Regularized geometric quantiles and universal linear distribution functionals Domain: math.ST Source: arxiv Authors: Dimitri Konen, Gilles Stupfler Reviews: 0 Subject: math.ST Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-02-10 Page: https://delphi.science/r/arxiv/2602.09356 Geometric quantiles are popular location functionals to build rank-based statistical procedures in multivariate settings. They are obtained through the minimization of a non-smooth convex objective function. As a result, the singularity of the directional derivatives leads to numerical instabilities and poor sample properties as well as surprising `phase transitions' from empirical to population distributions. To solve these issues, we introduce a regularized version of geometric distribution functions and quantiles that are provably close to the usual geometric concepts and share their…