Séminaire MODAL'X : Maxime Egéa (LPSM)

Résumé : Survival analysis is often applied to sensitive data making privacy protection of crucial importance. In this talk, we consider local differential privacy in survival analysis using private failure indicators. First, I will provide an overview of existing methods and discuss how individual privacy can be preserved.  Next, I will introduce a non-parametric kernel estimator for the cumulative hazard function using private mechanisms. Under mild conditions, we prove lower bounds on the minimax rates of convergence and show that our estimator is minimax optimal under well-chosen bandwidths. Finally, I will illustrate our estimator with numerical results.