报告时间：2016年11月22日（星期二）上午10：00-11：00
报告地点：管理学院1010会议室
报 告 人：张大力 博士
题目：Varying confidence levels for CVaR risk measures and minimax limits
摘要：Conditional value at risk (CVaR) has been widely used as a risk measure in finance. In this paper our focus is on the choice of confidence level. When the confidence level is set close to 1, the CVaR risk measure approximates the extreme (worst scenario) risk measure. We analyse optimization problems involving the CVaR risk measure, emphasizing problems with a risk constraint. We give conditions under which the optimal solution as the confidence level approaches 1 is well-behaved and approaches its natural limit. We also study the sample average approximation scheme for the CVaR constraints and investigate the convergence of the optimal solution obtained from this scheme as the sample size increases. Finally we have looked at the possibility that varying the confidence level to a lower value could give an advantage when there is a need to find good solutions to risk-constrained problems out of sample. We use some portfolio optimization problems to investigate a procedure which solves an adjusted problem with lower confidience level and a correspondingly higher value for the upper bound on risk. Our numerical results demonstrate that using the optimal solution to this adjusted problem can lead to better overall performance.
报告人简介：张大力博士：现任职于上海交通大学安泰经济与管理学院中美物流研究院助理教授，硕士生导师。2010年毕业于英国南安普顿大学数学学院/管理学院。2010年-2011年任于新加坡国立大学工业工程系博士后研究员，2011年-2013年任新加坡健保集团（国立，Singapore Health Service Ltd）医疗服务研究中心的政策分析师,主要负责相关国立医院(如新加坡中央医院、新加坡竹脚妇幼医院、新加坡眼科中心等)部门的流程建模与医疗资源管理项目。2013年至今，在上海交通大学工作，获得上海市浦江人才项目支持。2013年-2014年学术访问南安普顿大学数学学院/管理学院和麦吉尔大学管理学院/计算机学院。
