皇家华人

风险管理前沿系列讲座

时间：2025年6月8日9:00-17:00

地点：皇家华人 皇家华人A0406室

1）报告题目：Diversification for infinite-mean heavy-tailed losses without risk aversion

报告人：胡太忠（中国科学技术大学） 

摘要：In this talk, we introduce the portfolio diversification of infinite mean models. With the notion of majorization order, it is shown that a more diversified portfolio of iid ultra heavy-tailed Pareto losses is larger in the sense of the first-order stochastic dominance. This result is generalized to ultra heavy-tailed Pareto losses which are triggered by catastrophic events, and random losses of which the tails follow an ultra heavy-tailed Pareto distribution. Finally, we introduce a new class of distributions that includes many commonly used heavy-tailed distributions and show that within this class, the above property holds.Remarkably, different from the finite-mean setting, such a diversification benefit does not depend on the decision maker's risk aversion.

This is a joint work with with Y. Chen, S. Shneer, R. Wang and Z. Zou.

报告人简介：胡太忠，中国科学技术大学管理皇家华人教授，1994年于中国科大获概率论与数理统计专业博士学位，2004年入选教育部“新世纪优秀人才支持计划”，2014年获安徽省教学名师称号，2017年获宝钢教育基金优秀教师奖。研究领域包括精算理论、随机比较、统计相依、极值理论和可靠性数学等。现为Insurance: Mathematics & Economics, Probability in the Engineering and Informational Sciences, Quality Technology & Quantitative Management等国际期刊副主编，《中国科学技术大学学报》学科主编, 中国工业与应用数学“金融数学和工程专业委员会”副主任委员，中国现场统计研究会风险管理与精算分会副理事长。

2）报告题目：Statistical Modeling of Data Breach Lifecycle

报告人：赵鹏（江苏师范大学）

摘要：Data breache4s cause significant financial losses and have become a major concern for businesses. To mitigate the damage caused by a data breach, a key concept of the data breach lifecycle that involves three components, the occurrence of a breach, the time to detect the breach, and the time to report the breach, has to be well understood. In this talk, we initialize the statistical modeling of the data breach lifecycle via a self-exciting marked point process. The proposed model accommodates the heterogeneity between hacking and non-hacking events, and the dependence between two marks - the time to detect the breach and the time to report the breach - is modeled via a copula approach. The missing and censoring mechanisms are taken into account in the modeling process as well. Empirical studies show that the proposed approach has satisfactory fitting and predictive performance.

报告人简介：赵鹏，江苏师范大学教授、副校长。主要从事可靠性统计和网络可靠性等领域的研究工作。先后入选基金委国家优青项目、杰青项目。担任期刊Commun. Stat.副主编（AE）、《应用概率统计》《数理统计与管理》编委。曾获江苏省“双创人才”、江苏省数学成就奖、江苏省教学成果一等奖（第一完成人）等。担任中国现场统计研究会大数据统计分会副理事长、中国数学会理事、概率统计分会常务理事等。 

3）报告题目：Semiparametric Tests for Lorenz Dominance Based on Density Ratio Model

报告人：庄玮玮（中国科学技术大学）

摘要：Lorenz dominance is a fundamental tool for judging whether wealth disparity expands further in one country or region than another. Based on the popular density ratio model, we propose a new semiparametric test for Lorenz dominance. We show that the limiting distribution of the propose test statistic is a supremum of a Gaussian process. For practical use, we further devise a bootstrap procedure to calculate the p-value of our test and establish it validity. Our simulations show that our test not only has a valid type I error, but also outperforms its competitors in terms of power. Finally, we apply the proposed test to study the salaries of compare the individual salaries among higher education employees in Ohio from 2011-2015.

报告人简介：庄玮玮, 女，中国科学技术大学副教授，博士生导师，安徽省杰青项目获得者。主要研究方向包括随机比较与统计检验、风险管理等，已在Biometrika，Insur. Math. Econ.，J. Multivariate Anal.，J. Appl. Probab.等国际知名SCI期刊上发表多篇高质量论文，有多篇论文被Elsevier，Springer 出版社出版的专著收录。已主持完成3项国家级自然科学基金科研项目，包括面上项目 2 项、青年项目 1 项。曾获安徽省“自然科学论文一等奖”，安徽省“省级教坛新秀”，并以主要成员获国家级线下一流课程，安徽省教学成果特等奖，安徽省教学成果一等奖等。