Publications

La plupart des informations présentées ci-dessous ont été récupérées via RePEc avec l'aimable autorisation de Christian Zimmermann
Machine learning for credit scoring: Improving logistic regression with non-linear decision-tree effectsJournal articleElena Dumitrescu, Sullivan Hué, Christophe Hurlin et Sessi Tokpavi, European Journal of Operational Research, Volume 297, Issue 3, pp. 1178-1192, 2022

In the context of credit scoring, ensemble methods based on decision trees, such as the random forest method, provide better classification performance than standard logistic regression models. However, logistic regression remains the benchmark in the credit risk industry mainly because the lack of interpretability of ensemble methods is incompatible with the requirements of financial regulators. In this paper, we propose a high-performance and interpretable credit scoring method called penalised logistic tree regression (PLTR), which uses information from decision trees to improve the performance of logistic regression. Formally, rules extracted from various short-depth decision trees built with original predictive variables are used as predictors in a penalised logistic regression model. PLTR allows us to capture non-linear effects that can arise in credit scoring data while preserving the intrinsic interpretability of the logistic regression model. Monte Carlo simulations and empirical applications using four real credit default datasets show that PLTR predicts credit risk significantly more accurately than logistic regression and compares competitively to the random forest method.

Social Ties and the Influence of Public Policies on Individual Opinions: The Case of Same-Sex Marriage LawsJournal articleSylvie Blasco, Eva Moreno Galbis et Jérémy Tanguy, The Journal of Law, Economics, and Organization, Volume 38, Issue 1, pp. 196-271, 2022

This paper evaluates if same-sex marriage (SSM) laws, approved in several European Union countries over the past decades, have contributed to favor gay-friendly opinions among people depending on their social interactions. We propose a dyadic model in which individuals learn about the social norm conveyed by a law through strong and weak ties. We show that the relative importance of these social ties in shaping individuals’ opinions depends on the alignment between the law and the local social norm. Using the 2002–2016 European Social Surveys, we test the theoretical predictions with a pseudo-panel dynamic difference-in-difference setting relying on the progressive adoption of SSM in European countries. We show that strong ties induce a lower increase in gay-friendly opinions following the adoption of SSM when the law is aligned with the local social norm. When the law clashes with this norm, strong ties induce a larger increase.

The French school of proximity – Genesis and evolution of a school of thoughtBook chapterJean-Benoît Zimmermann, André Torre et Michel Grossetti, In: Handbook of Proximity Relations, A. Torre et D. Galaud (Eds.), 2022-01, pp. 49-69, 2022

In this chapter, we revisit the origins and genesis of the french school of proximity and its evolution trough time, in order to better understand how and why the small group of researchers who were the driving force of this new way of thinking were quickly able to get a real legitimacy and effective recognition. First of all, it was clear that the role of space in economic dynamics was too often the subject of confusion and abusive assertions. Asking this question in terms of coordination made it possible to consider non-spatial factors in the analysis. The notion of proximity as a polysemic concept therefore opened the way to understanding how space matters or not, together with these other factors thus a renewed approach of questions related to space and territories. But, even starting from issues of economic nature, such an approach could not remain limited to its economic dimension, the questions of coordination involving social individuals, located in geographical space but also embedded in bundles of relationships and in institutions. Thus, it had to broaden very quickly to other disciplines in social sciences which largely contributed to consolidate the bases of what became a multidisciplinary approach and to develop theoretical as well as empirical tools.