时间：5月19日（周六），上午9：00 – 12：00
地点：明德主楼1037会议室
报告1：Stochastic Distortion and its Transformed Copula
报告人：杨静平，北京大学数学科学学院金融数学系教授，博士生导师。现任数量经济与数理金融教育部重点实验室（北京大学）副主任，中国工业与应用数学学会第七届理事会理事。研究兴趣有金融和保险中的风险相依性、债券组合模型和信贷资产证券化等。
报告摘要：Motivated by wide applications of distortion functions and copulas in insurance and finance, we generalize the notion of a deterministic distortion function to a stochastic distortion, i.e., a random process, and employ the defined stochastic distortion to construct a so-called transformed copula by stochastic distortions. One method for constructing stochastic distortions is provided with a focus on using time-changed processes. After giving some families of the transformed copulas by stochastic distortions, a particular class of transformed copulas is applied to a portfolio credit risk model, where a numeric study shows the advantage of using the transformed copulas over the conventional Gaussian copula and the double t copula in terms of the fitting accuracy and the ability of catching tail dependence. It is a joint work with Feng Lin, Liang Peng and Jiehua Xie.
报告2：New Progress in Machine Learning and Actuarial Modeling
报告人：张连增，南开大学金融学院精算学系教授，博士生导师。先后在南开大学经济学院风险管理与保险学系、南开大学金融学院任教，曾是墨尔本大学、滑铁卢大学、洛桑大学访问学者。主要从事精算风险理论、非寿险统计建模、机器学习精算应用等方面的研究。
报告摘要：In this talk, I provide a brief review of machine learning in actuarial predictive modeling. In particular, I illustrate neural networks with two examples, and make a comparison with the usual linear regression models in the first example. Further research can be done to discuss applications of neural networks in classification. One key technique in neural networks is the gradient descent method. A related method is Newton’s method. I provide a brief comparison from the definition, which may be helpful for the beginners.
报告3：Big Data, AI and Statistical Thinking
报告人：刘乐平，天津财经大学统计学、金融学教授，博士生导师。现任天津财经大学大数据统计研究中心主任，在研国家自然科学基金项目“基于机器学习的长期护理保险精算预测模型与风险分析”。
报告摘要：If big data is the “ocean”, then statistics must be one of its main streams. Statistics have produced many important scientific discoveries in the historical process of human’s inquiry into the uncertainties of the world. As the wave of big data surges, reviewing the history of statistical development, drawing wise statistical thinking and using these thoughts to question big data and artificial intelligence, may help (p <0.05) statisticians better grasp the future direction in the big data ocean.
报告4：Optimal Insurance Contracts with Background Risk and Higher-order Risk Attitudes
报告人：池义春，中央财经大学中国精算研究院研究员。主要从事风险理论、最优保险/再保险设计以及变额年金的定价和对冲等方面研究，在国际四大精算学期刊上发表了二十篇学术论文。2012年荣获北美产险精算学会Hachemeister奖，2015年破格晋升为研究员。
报告摘要：In this talk, we discuss an optimal insurance problem in the presence of background risk from the perspective of an insured with higher-order risk attitudes. We introduce several useful dependence notions to model positive dependence structures between the insurable risk and background risk. Under these dependence structures, we compare insurance contracts of different forms in higher-order risk attitudes and establish the optimality of stop-loss insurance form. We also explicitly derive the optimal retention level. Finally, we carry out a comparative analysis and investigate how the change in the insured’s initial wealth or background risk affects the optimal retention level. This is a joint work with Wei Wei.