Finding local vote transfer matrices with an inverse method

Read full text Summarise with VIRAT
Authors: Pablo Jensen, Nelly Pustelnik, Rémi Gribonval, Pascal Carrivain
Cited by: 0
References: 9

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

We propose an original approach, based on inverse methods, to compute local vote transfer matrices. Our method defines a similarity metric that allows to define neighbors for each precinct, and assumes that neighbor precincts have similar transfer matrices. We analyze the two rounds of the 2017 French presidential election, and recover the aggregated transfer matrix with a single free parameter. We further show that transfers to the extreme-right candidate Le Pen depend on its score in the first round, and that transfers to abstention from the right candidate Fillon depend on the percentage of non voters in the first round.

Fields of Study & Subjects

Electoral Systems and Political Participation Internet Traffic Analysis and Secure E-voting Benford’s Law and Fraud Detection

Publication details

Journal / source
HAL
Publication date
25 March 2026
Language
EN
Document type
Preprint