Home > Research > Publications & Outputs > A probabilistic evaluation framework for prefer...
View graph of relations

A probabilistic evaluation framework for preference aggregation reflecting group homogeneity

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published
Close
<mark>Journal publication date</mark>30/11/2018
<mark>Journal</mark>Mathematical Social Sciences
Volume96
Number of pages14
Pages (from-to)49-62
Publication StatusPublished
Early online date27/09/18
<mark>Original language</mark>English

Abstract

Groups differ in the homogeneity of their members’ preferences. Reflecting this, we propose a probabilistic criterion for evaluating and comparing the adequateness of preference aggregation procedures that takes into account information on the considered group’s homogeneity structure. Further, we discuss two approaches for approximating our criterion if information is only imperfectly given and show how to estimate these approximations from data. As a preparation, we elaborate some general minimal requirements for measuring homogeneity and discuss a specific proposal for a homogeneity measure. Finally, we investigate our framework by comparing aggregation rules in a simulation study.