Model‐Agnostic Influential Outlier Metric
Stat · 2026-09 · 15 (3)
DOI: 10.1002/sta4.70163 ↗Kurz gefasst
Fachartikel in der Zeitschrift Stat, online erschienen am 07.07.2026. Verfasst von Colin C. Jones und 1 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The influential outlier metric (IOM) provides model‐agnostic influential outlier detection. We define influence of an observation using a combination of SHapley Additive exPlanation (SHAP) values and the residual. Both are transformed using normalizing flows, changing their respective measures to Gaussian distributions. The transformed SHAP values and residuals define the IOM as a product of two independent random variables, allowing critical values for a given significance level . We illustrate the advantages and limitations of this approach in several models including the linear model, neural networks, the random forest and gradient‐boosted trees.
Zitieren
APA
Jones, C. C., & Campbell, D. A. (2026). Model‐Agnostic Influential Outlier Metric. Stat, 15(3). https://doi.org/10.1002/sta4.70163
Vancouver
Jones CC, Campbell DA. Model‐Agnostic Influential Outlier Metric. Stat. 2026;15(3). doi:10.1002/sta4.70163
BibTeX
@article{Jones2026,
author = {Jones, Colin C. and Campbell, David A.},
title = {Model‐Agnostic Influential Outlier Metric},
journal = {Stat},
year = {2026},
volume = {15},
number = {3},
doi = {10.1002/sta4.70163}
}
EUVIMED