Abstract: The model uncertainty is of crucial importance when market participants are making risk management strategies. For a participant who adopts law-invariant risk measures for quantification, the study of the supremum of risk measure values can help the participant to better understand the performance of risk in the worst-case scenario. In this talk, we introduce several model uncertainty settings. The choices of risk measures, uncertainty sets, and transformations of the underlying risk play important roles in the characterization of the worst-case distribution. Motivated by the insurance policies, we mainly focus on stop-loss functions and limited loss functions. Furthermore, we discuss the optimal retention levels for participants in an insurance policy with model uncertainties.

Bio: Dr. Fangda Liu is an assistant professor at the Department of Statistics and Actuarial Science, University of Waterloo. She is also an associated fellow of the Society of Actuaries. Her research focuses on optimal insurance and reinsurance design, risk sharing problems, distributional model uncertainty and its application insurance. 

Tencent:
https://meeting.tencent.com/dm/BizlE9URPqMC
ID: 607-128-288