Time: 16:00, August 29, 2019
Location: Room 506, Business School
Brief Introduction of the Speaker:
Xu Fengmin, female, Zhengzhou, Henan Province, Doctor of Computational Mathematics, Professor of School of Economics and Finance, Xi'an Jiaotong University, Doctoral Tutor, Assistant Dean and Head of Department of Financial Engineering. Visiting Scholar of Simon Fraser University, Canada, Visiting Scholar of Hong Kong Polytechnic University, Senior Exchange Scholar of Seoul National University, Korea, Winner of Shaanxi Youths’ Science and Technology Award, Director of China Double Election Law Society, Vice President and Secretary-General of Economic Mathematics and Management Mathematics Branch of China Double Election Law Society, Standing Director of Mathematics Planning Branch of Operation Research Society of China.
Applicants have long been committed to the research of statistical and sparse optimization theory and algorithms involved in big data and micro-research of typical financial problems. He published many articles in well-known periodicals at home and abroad. He was involved in the compilation of a monograph. He presided over two projects of the National Natural Science Foundation and joined in a key project of the Natural Science Foundation.
In the practical business environment, portfolio managers often face business-driven requirements that limit the number of constituents in their optimal portfolio. A natural sparse finance optimization model is thus to minimize a given objective function while enforcing an upper bound on the number of assets in the portfolio. In this talk we consider why we select sparse financial model and how to select the optimal sparsity parameter. Furthermore, Sparse and group sparse index tracking models and algorithms are presented, and we conduct empirical tests to demonstrate that our approach generally produces sparse portfolios with higher out-of –sample tracking error.