教育部人文社科重点研究基地中央财经大学中国精算研究院学术活动

精算论坛第130期讲座-Vali Asimit (3月21日)

报告题目：An efficient approach to quantilecapital allocation and sensitivity analysis  

报告时间：2018年3月21日  上午10:30—11:30

报告人：ValiAsimit 教授（CassBusiness School, City University of London）

摘要：

Invarious fields of applications such as capital allocation, sensitivity analysisand systemic risk evaluation, one often needs to compute or estimate theexpectation of a random variable given that another random variable is equal toits quantile at some pre-specified probability level. A primary example of suchan application is the Euler capital allocation formula for the quantile (oftencalled the Value-at-Risk), which is of crucial importance in financial riskmanagement. It is well known that classic nonparametric estimation for theabove quantile allocation problem has a slower rate of convergence than thestandard rate. In this paper, we propose an alternative approach to thequantile allocation problem via an approximation based on another conditionalexpectation. The two methodologies often give identical or asymptoticallyequivalent values, and the new approach enjoys some nice robustness properties.The asymptotic distribution of the proposed nonparametric estimator of the newcapital allocation is derived for dependent data, namely under the setup ofmixing sequences. In order to assess the performance of the proposed nonparametricestimator, AR-GARCH models are proposed to fit each risk variable, and furthera bootstrap method is employed to quantify the estimation uncertainty. Asimulation study is conducted to examine the finite sample performance of theproposed inference. Finally, the proposed methodology of quantile capitalallocation is illustrated for a financial data set.  

报告地点：中央财经大学学术会堂南楼506（精算院会议室）

欢迎各位老师和同学积极参加！