一、

报告题目：On some sampling distributions for skew normal population and its applications

报告时间：2016年8月2日 上午10：50-11:35

报告地点：万象城官方网站（中国）有限公司504室

报告人:苏南城 副教授

Abstract ：There are intense investigations about generalizing normal distributions. Among them, the skew-normal distribution retains many statistical properties of the normal distribution. In this talk, first some basic properties of the skew-normal distribution and its related research are reviewed. Next, we study properties of sampling distributions for skew-normal population, including explicit forms of the probability density function and cumulative density function of the sample mean, and the joint moment generating function of sample mean and sample variance, etc. Finally, we illustrate the distribution of skew-normal sample mean with an application to control charts with skewed measurements.

报告人简介：

苏南城为台北大学统计系副教授，Journal of Data Science技术编辑，研究方向：次序与记录统计、特征分布、广义分布理论与应用、质量控制，已发表学术论文20多篇，获得台北大学优秀学者奖。

二、

报告题目：A Linearization of Portfolio Optimization Problem with General Risk Measures under Multivariate Conditional Heteroscedastic Models

报告时间：2016年8月2日上午10:00-10:45

报告地点：万象城官方网站（中国）有限公司504室

报 告 人：黄士峰 教授

Abstract：This study considers a portfolio selection problem with general risk measures, where the dynamics of the underlying returns are described by multivariate autocorrelated and conditional heteroscedastic models. A linearlization of the portfolio optimization problem is proposed to obtain the optimal asset allocations by linear programming. The linearization representations of the problem with expected-shortfall and spectral risk measure are derived. Simulation results indicate that the proposed scheme saves computational time remarkably and reveal that autocorrelation and conditional heteroscedasticity inherent in return processes cannot be ignored in investment decision-making. An empirical study is conducted with the stocks of the FTSE TWSE Taiwan Mid-Cap 100 Index. A self-financing trading strategy is established by reallocating the holding positions via the proposed scheme daily. Numerical results indicate that transaction costs and tax reduce the performance of the self-financing trading strategy substantially, unless the brokerage companies offer low discounted transaction fees.

报告人简介：

黄士峰现为台湾高雄大学应用数学系教授，统计研究所所长，大数据研究中心主任。主要研究方向为时间序列、金融工程、金融统计、风险管理， 在Annals of Statistics、Quantitative Finance等杂志发表20多篇高水平论文，获得魏庆荣统计论文奖，高雄大学优秀学者奖等。