91直播

91直播：杨洋 教授 (南京审计大学)

时间：2024年11月22日 14:00-

地点: 数统91直播 LD302

摘要：With recent developments of systemic risk across the banking, finance, and insurance sectors, this paper examines three higher-order tail moments, i.e. the generalized marginal/systemic/joint expected shortfall (GMES/GSES/GJES), in a context of quantitative risk management, associated with a confidence level q∈(0, 1). Our analysis involves a system encompassing multiple sub-portfolios, business lines, or entities, in which the loss-profit variables are modeled by randomly weighted random variables with heavy tails and two types of tail dependence. Specifically, for heavy-tailed net losses, whether asymptotically independent or asymptotically dependent, we derive several general asymptotic formulas for the GMES, GSES, and GJES of the aggregate discounted loss as q approaches 1, which help to better understand systemic risk and its measurements in the financial domain. We further refine the asymptotic results we obtained under the expected shortfall capital allocation rule. Our main finding is that the asymptotic independence among net loss variables has no contribution to the above three risk measures, while the asymptotic dependence does contribute.

邀请人: 张志民

欢迎广大师生积极参与！