Distributional Robustness and Uncertainty Quantification for Rare Events

arXiv1911.09580MaRDI QIDQ6329641FDOQ6329641

Authors: Jeremiah Birrell, Paul Dupuis, Markos Katsoulakis, Luc Rey-Bellet, Jie Wang 

Publication date: 21 November 2019

Abstract: Rare events, and more general risk-sensitive quantities-of-interest (QoIs), are significantly impacted by uncertainty in the tail behavior of a distribution. Uncertainty in the tail can take many different forms, each of which leads to a particular ambiguity set of alternative models. Distributional robustness bounds over such an ambiguity set constitute a stress-test of the model. In this paper we develop a method, utilizing R'enyi-divergences, of constructing the ambiguity set that captures a user-specified form of tail-perturbation. We then obtain distributional robustness bounds (performance guarantees) for risk-sensitive QoIs over these ambiguity sets, using the known connection between R'enyi-divergences and robustness for risk-sensitive QoIs. We also expand on this connection in several ways, including a generalization of the Donsker-Varadhan variational formula to R'enyi divergences, and various tightness results. These ideas are illustrated through applications to uncertainty quantification in a model of lithium-ion battery failure, robustness of large deviations rate functions, and risk-sensitive distributionally robust optimization for option pricing.

This page was built for publication: Distributional Robustness and Uncertainty Quantification for Rare Events

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6329641)