Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Acemoglu, Sinan, Kleiber, Christian, Urban, Jörg
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2601.00773
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909980053995520
author Acemoglu, Sinan
Kleiber, Christian
Urban, Jörg
author_facet Acemoglu, Sinan
Kleiber, Christian
Urban, Jörg
contents Variable importance in regression analyses is of considerable interest in a variety of fields. There is no unique method for assessing variable importance. However, a substantial share of the available literature employs Shapley values, either explicitly or implicitly, to decompose a suitable goodness-of-fit measure, in the linear regression model typically the classical $R^2$. Beyond linear regression, there is no generally accepted goodness-of-fit measure, only a variety of pseudo-$R^2$s. We formulate and discuss the desirable properties of goodness-of-fit measures that enable Shapley values to be interpreted in terms of relative, and even absolute, importance. We suggest to use a pseudo-$R^2$ based on the Kullback-Leibler divergence, the Kullback-Leibler $R^2$, which has a convenient form for generalized linear models and permits to unify and extend previous work on variable importance for linear and nonlinear models. Several examples are presented, using data from public health and insurance.
format Preprint
id arxiv_https___arxiv_org_abs_2601_00773
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Variable Importance in Generalized Linear Models -- A Unifying View Using Shapley Values
Acemoglu, Sinan
Kleiber, Christian
Urban, Jörg
Methodology
Variable importance in regression analyses is of considerable interest in a variety of fields. There is no unique method for assessing variable importance. However, a substantial share of the available literature employs Shapley values, either explicitly or implicitly, to decompose a suitable goodness-of-fit measure, in the linear regression model typically the classical $R^2$. Beyond linear regression, there is no generally accepted goodness-of-fit measure, only a variety of pseudo-$R^2$s. We formulate and discuss the desirable properties of goodness-of-fit measures that enable Shapley values to be interpreted in terms of relative, and even absolute, importance. We suggest to use a pseudo-$R^2$ based on the Kullback-Leibler divergence, the Kullback-Leibler $R^2$, which has a convenient form for generalized linear models and permits to unify and extend previous work on variable importance for linear and nonlinear models. Several examples are presented, using data from public health and insurance.
title Variable Importance in Generalized Linear Models -- A Unifying View Using Shapley Values
topic Methodology
url https://arxiv.org/abs/2601.00773