Germain Gauthier

Assistant Professor of Economics
Department of Social and Political Sciences
Bocconi University

germain.gauthier@unibocconi.it
curriculum vitae

Research

Methods

Among other things, I am developing new AI methods for social scientists working with unstructured data, such as texts, images, audio, and video recordings.

  • Measuring Crime Incidence and Reporting: Method and Application to #MeToo
    working paper

  • Deep Latent Variable Models for Unstructured Data
    (with Philine Widmer & Elliott Ash)
    Paper available upon request

  • Neural Ideal Point Models
    (with Hugo Subtil & Philine Widmer)
    Paper available upon request

  • Relatio: Text Semantics Capture Political and Economic Narratives
    (with Elliott Ash & Philine Widmer)
    Political Analysis, 2023
    published version, working paper, replication files

Applied Work

I mainly study the societal consequences of digital technologies, such as social media and artificial intelligence.

  • The Lifecycle of Protests in the Digital Age
    (with Pierre Boyer, Yves Le Yaouanq, Vincent Rollet & Benoît Schmutz)
    working paper

  • The Political Effects of X’s Recommender Algorithm
    (with Roland Hodler, Philine Widmer & Ekaterina Zhuravskaya)
    Swiss National Science Foundation Grant No. 100018_215554
    Paper available upon request

  • Artificial Inequality: Large Language Models Disadvantage Women Job Seekers
    (with Kenza Elass, Debora Nozza, and Paola Profeta)
    Paper available upon request

  • Les déterminants de la mobilisation des “Gilets Jaunes”
    (with Pierre Boyer, Thomas Delemotte, Vincent Rollet & Benoît Schmutz)
    Revue Economique, Vol. 71 - 1. Jan. 2020
    published version, policy brief, column

Teaching

  • Graduate Level - Diversity and Global Policy (Bocconi, 2023/2024)
    slides

  • Undergraduate Level - Public Finance (Bocconi, 2023/2024)
    slides

  • Graduate Level - Text as Data for the Social Sciences (LMU, 2022/2023)
    slides

  • Graduate Level - Introductory Course to Stata (CREST, 2020)
    gitlab

Software

  • relatio: A Python package to extract narrative statements from texts
    github, pypi

  • gtm: A Python package to estimate flexible neural topic models
    github

  • IdealPointNN: A Python package to estimate neural ideal point models
    github