Multi-objective approaches to portfolio optimization with market impact costs

doi:10.1007/s12293-022-00381-w

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 机器人理论与应用

	Multi-objective approaches to portfolio optimization with market impact costs
	Wang, Hongze1,2 ; Li, Xuerong 3; Hong, Wenjing 4; Tang, Ke 4
发表期刊	MEMETIC COMPUTING
ISSN	1865-9284
	2022-10-22
页码	11
通讯作者	Li, Xuerong(lixuerong@amss.ac.cn)
摘要	Market impact costs are important factors to portfolio management, which always lead to adverse price fluctuations in trading. As the practical trading volume becomes increasingly large, the Problem of Portfolio Optimization with Market Impact Costs (MICPOP) has become more important. Traditional MICPOPs involve seeking an optimal allocation of capital to a limited number of assets with respect to either additional constraints or a weighted sum of objectives (net return, investment risk). We suggest solving MICPOPs with Multi-Objective Evolutionary Algorithms (MOEAs). Specifically, we formulate MICPOPs as a bi-objective optimization problem. The advantages of MOEAs over state-of-the-art single objective approaches to MICPOPs will be shown through empirical studies. Our study has revealed that a well-known MOEA, namely Nondominated Sorting Genetic Algorithm II (NSGA-II), fails to provide satisfactory solution quality sometimes. Hence, a memetic MOEA for Portfolio Optimization with Market Impact costs (POMI-MOEA), which inherits the global search capability of NSGA-II while introducing a new local search operator, is proposed and evaluated in this paper. Comprehensive experimental studies on 11 portfolio cases have shown the superiority of POMI-MOEA over NSGA-II and other two MOEAs for MICPOPs.
关键词	Portfolio optimization Market impact cost Multi-objective optimization Evolutionary computation Memetic algorithm
DOI	10.1007/s12293-022-00381-w
关键词[WOS]	TRANSACTION COST ; ALGORITHM ; MODEL
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[71901205]
项目资助者	National Natural Science Foundation of China
WOS研究方向	Computer Science ; Operations Research & Management Science
WOS类目	Computer Science, Artificial Intelligence ; Operations Research & Management Science
WOS记录号	WOS:000870949700001
出版者	SPRINGER HEIDELBERG
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/50264
专题	多模态人工智能系统全国重点实验室_机器人理论与应用
通讯作者	Li, Xuerong
作者单位	1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 3.Chinese Acad Sci, Acad Math & Syst Sci, Beijing 100190, Peoples R China 4.Southern Univ Sci & Technol, Dept Comp Sci & Engn, Guangdong Prov Key Lab Brain Inspired Intelligent, Shenzhen 518055, Peoples R China
第一作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Wang, Hongze,Li, Xuerong,Hong, Wenjing,et al. Multi-objective approaches to portfolio optimization with market impact costs[J]. MEMETIC COMPUTING,2022:11.
APA	Wang, Hongze,Li, Xuerong,Hong, Wenjing,&Tang, Ke.(2022).Multi-objective approaches to portfolio optimization with market impact costs.MEMETIC COMPUTING,11.
MLA	Wang, Hongze,et al."Multi-objective approaches to portfolio optimization with market impact costs".MEMETIC COMPUTING (2022):11.