周知，重尾模型得到广泛的研究与应用，已经成为应用概率统计的一个重要的研究领域。为进一步提高我国重尾模型的理论与应用的研究水平，加强同行之间的交流与合作，经中国概率统计学会及威尼斯人-威尼斯人娱乐场 批准，在统计学江苏省一级学科重点项目资助下，继2010年第一届重尾模型及其应用研讨会之后，我们将于2014年3月21日-23日，在威尼斯人-威尼斯人娱乐场 召开“第二届重尾模型及其应用研讨会”，以促进本方向研究和应用的发展。届时，英国Heriot-Watt University的Foss Sergey教授，美国University of Iowa的Tang Qihe教授将应邀与会做中心报告。
 
 
 
 
 
时间：2014年3月22日（星期六）9:00
地点：数学楼学术报告厅 、东吴饭店

报告一题目：An overview on some results and open problems related to heavy-tailed distributions
报告人：Prof. Sergey Foss（Heriot-Watt University, Edinburgh；S.L.Sobolev Institute of Mathematics, Novosibirsk）
报告摘要：I will speak about various topics related to heavy tails, formulate some results and hypotheses, and introduce a number of open problems.
I will start with asymptotics analysis of tail probabilities for supremum $M=sup_n S_n$ of random walk $S_0=0, S_n=X_1+...+X_n$ given $M$ is finite a.s. and review the 5 main cases, two cases for heavy-tailed distributions and three cases for light-tailed distributions. Then I formulate open questions related to the "intermediate" light-tailed case. This relates to the papers [DFK] and [FP].
Second, I will consider sums and maxima of dependent random variables in the presence of heavy tails, discuss general assumptions related to "conditional independence" and provide examples. This part is based on the paper [FR].

Third, I will talk about the distributional asymptotics for time and value of an overshoot over a high level, for a modulated random walk with heavy-tailed distributions. This related to the paper [AF] and to the book [FKZ] (second edition).
If time allows, I'll also talk about a number of open problems on the tail asymptotics for a number of characteristics of multi-dimensional stochastic queues and networks. This part relates to the papers [BF], [FK1], [FK2] and [FM].

报告二题目：The Sum-Product Structure as a Mechanism for Risk Management
报告人：Prof. Qihe Tang（Department of Statistics and Actuarial Science, University of Iowa）
报告摘要：The sum-product structure $sum_{i=1}^{n}X_{i}prod_{j=1}^{i}Y_{j}$ for $n$ finite or infinite appears naturally in insurance and finance modeling, where the real-valued $X$ random variables and the positive $Y$ random variables are often interpreted as insurance risks and financial risks, respectively. This talk will demonstrate its versatility for risk management. A few new results for the heavy-tailed case will be shown.
 
时间：2014年3月23日（星期日）9:00
地点：数学楼学术报告厅、东吴饭店

报告一题目：How could "heavy tails" appear in the "light-tailed" world
报告人：Prof. Sergey Foss（Heriot-Watt University, Edinburgh； Institute of Mathematics, Novosibirsk）
报告摘要：Poisson and normal distributions are the two famous fundamental distributions in probability and statistics. Both are "light-tailed", i.e. have finite exponential moments. I will discuss two examples on how could they lead naturally to heavy-tailed (namely Weibull-type with parameter <1) distributions. First example deals with products of normal random variables, second with Poisson processes on the plane modelling some telecommunication systems.

报告二题目：Versatility of Extreme Value Theory in Insurance and Finance
报告人：Prof. Qihe Tang（Department of Statistics and Actuarial Science, University of Iowa）
报告摘要：The prevalence of Black-Swan events accompanied by disastrous economic and social consequences makes today's world far different from just decades ago.  It intensifies the urgent need for quantitatively understanding extreme risks in the insurance and financial industry.  In this talk, after a brief introduction of extreme value theory I shall illustrate its versatility in quantitative risk management by presenting its applications to various topics in the interdisciplinary area of insurance and finance.