Institution profile

Institut National de la Santé et de la Recherche Médicale

Academic institutioneurope · fr
Official website
Research library275linked papers
Opportunities0open roles
Selected work

Representative Papers

Rethinking the Win Ratio: A Causal Framework for Hierarchical Outcome Analysis

Jan 28, 2025

Conventional methods such as the Win Ratio lack a rigorous statistical and causal foundation for quantifying treatment effects in hierarchical multivariate outcomes, potentially leading to erroneous treatment recommendations in heterogeneous populations. Method: We establish, for the first time, a formal causal framework for the Win Ratio—defining identifiable, individual-level causal estimands—and propose a novel estimation procedure based on paired nearest-neighbor matching and doubly robust estimation. Contribution/Results: We prove theoretical consistency and double robustness of the proposed estimator. Extensive validation on synthetic data and the CRASH-3 clinical trial demonstrates substantial improvements in reliability and interpretability of treatment effect estimation. Our work provides both a principled theoretical foundation and a practical methodology for causal inference with complex, hierarchically structured multiple endpoints.

2 citationsRead paper

Model Agnostic Differentially Private Causal Inference

May 26, 2025

Estimating the average treatment effect (ATE) under differential privacy from privacy-sensitive observational data—e.g., healthcare or economic records—faces challenges including strong modeling assumptions, privacy cost scaling with estimator complexity, and limited methodological flexibility. This paper proposes the first general framework that decouples nuisance parameter estimation from privacy protection via folded-splitting and ensemble prediction perturbation. Without assuming a specific data-generating mechanism or parametric model form, it delivers unified differential privacy guarantees for diverse ATE estimators—including G-formula, inverse probability weighting (IPW), and augmented IPW (AIPW). The framework further extends to differentially private meta-analysis of ATEs across multiple private data sources. We establish rigorous differential privacy and statistical efficiency guarantees. Empirical evaluation demonstrates that, under realistic privacy budgets, our method achieves ATE estimation accuracy close to non-private baselines while enabling robust cross-source result integration.

1 citationsRead paper
Recent publications

Latest Papers