报告题目: Sparse portfolio optimization with transaction costs
报告人:刘成(武汉大学教授)
报告时间:2024年3月14日16:00-17:00
报告地点:文波楼109智慧教室
摘要: This paper proposes a novel approach to construct the optimal sparse portfolio weight for the global minimum variance (GMV) portfolio. An appropriate objective function with two l1-norm penalties where the first penalty is on the product of the inverse of the covariance matrix and the vector of ones and the other penalty is on the turnover of the updated weight, is proposed to get a sparse portfolio weight and reduce the transaction costs effectively. We develop an efficient algorithm based on the alternating direction method of multipliers (ADMM) to solve the resultant constrained convex minimization problem. We show the advantages of our proposed approach through both theoretical and empirical analysis when the number of assets N and the sample size of asset returns T satisfying N/T → c for some c ∈ (0,∞). Extensive empirical studies show that the performance of our approach is superior to the plugging global minimum variance portfolio strategies in terms of lower out-of- sample volatility of portfolio returns and turnover of portfolio weights.
主讲人简介: 刘成，武汉大学经济与管理学院教授、博士生导师，入选国家级高层次人才青年项目。主要研究领域为金融计量经济学、理论计量经济学和数据科学。近五年以第一作者或通讯作者在统计学和计量经济学国际顶尖期刊Journal of the American Statistical Association 、Journal of Econometrics上发表文章多篇，相关成果也获得了湖北省人文社会科学优秀成果二等奖。主持国家自然科学基金面上项目、青年项目和教育部人文社会科学基金各1项。
