Quantitative methods and interdisciplinarity

A constant drive for methodological innovation sits at the heart of the department's identity. Our teams develop cutting-edge econometric and statistical tools to meet the challenges posed by large, complex or heterogeneous datasets. this includes causal modelling, robust methods for noisy or incomplete data, machine-learning algorithms, and gradient-free optimisation techniques. We combine scientific rigour with a strong interdisciplinary openness, building solid ties with the Institut Polytechnique de Paris - particularly with Hi! Paris and Energy 4 Climate - and with partners across France and abroad. These efforts feed into a high-level doctoral training, offered through the IP Paris doctoral school, which supports nearly 70 PhD students.

Econometric methods for applied research

Our endeavours in creating new econometric methods aim to address the limitations of current practices in applied research. This includes the development of new-difference-in-difference methods, tackling data combination issues, and exploring complex dependency and nonlinear panel data models. These contributions are essential for refining empirical research techniques across various disciplines.

Cross-disciplinary research and methodological innovation

Our unit's research spans the interplay betweekn machine learning and economics, the cost of fairness in online decision problems, and the impact of tax and environmental competition on firm behaviour. By exploring these and other topics, we contribute to the development of algorithms and analytical frameworks that address real-life problems, underscoring the importance of interdisciplinary approaches in tackling contemporary challenges.