Home » Profils » Godard Mathilde

PR[AI]RIE Researcher

Godard Mathilde

CNRS Researcher at Dauphine-PSL (Economics Department, LEDa).  Her research focuses on the effects of economic uncertainty on health, as well as on the interaction of social insurance programs with health and labor markets. She combines causal inference methods with natural experiments to answer causal questions, but is also interested in the recent developments of ML for causal inference. As affiliated professor at PSL, she teaches “Machine Learning for Economists” at Dauphine-PSL. Her research has been funded by the European Commission (Marie Curie IEF), and more recently by the ANR JCJC RecessionsHealth (2021). She has published articles in the Journal of Health Economics, Journal of Human Resources, AEJ: Economic Policy, and Health Economics She is also a member of ISNS (Institut Santé Numérique en Société) at PariSanté Campus. At ISNS, she coordinates a group working on the development of a simplified, shared relational schema for the SNDS, to facilitate code reuse across different research projects.

Informations

Project in Prairie

• Disciplinary research: She will continue to pursue research in health economics, taking advantage
of the incredible opportunity to use health claims data from the National Health Data System (SNDS). This data can now be linked with administrative registers with socio-economic info at the individual level (EDP-Santé (3.4 million individuals, 2008-2022)). She will pursue her work on the interplay between insecure careers and health over the life-course, combining causal inference methods with Machine Learning (ML) techniques and natural experiments. In particular, she will study the health spillovers of mass layoffs, the matching of patients to doctors, and the role of prescription style on health and labor-market trajectories.

• Interdisciplinary research: Despite its initial focus on causal inference, there’s a growing
interest in economics for ML and its applications (collect new data, prediction for policy, ML for
econometrics). Conversely, there’s a clear interest in ML for causal inference. At the intersection of
the two disciplines, there is a fast-growing literature on high-dimensionality and endogeneity, or on
the estimation of heterogeneous treatment effects for optimal resource allocation. Within PR[AI]RIE,
she is eager to collaborate with mathematicians or computer scientists with an interest in causal inference methods.