系统工程学报2024,Vol.39Issue(1) :48-60.DOI:10.13383/j.cnki.jse.2024.01.004

学习带边信息专家意见的在线投资组合策略

Online portfolio selection strategy by learning from expert advice with side information

杨兴雨 郑丽娜 林虹 黄帅
系统工程学报2024,Vol.39Issue(1) :48-60.DOI:10.13383/j.cnki.jse.2024.01.004

学习带边信息专家意见的在线投资组合策略

Online portfolio selection strategy by learning from expert advice with side information

杨兴雨 1郑丽娜 1林虹 1黄帅1
扫码查看

作者信息

  • 1. 广东工业大学管理学院,广东广州 510520
  • 折叠

摘要

针对以往学习专家意见的在线投资组合策略中专家策略并未考虑有助于提高投资者收益的边信息的不足,选取在相同边信息状态下投资相同单只股票、不同边信息状态下可能投资不同单只股票的策略为专家意见,基于指数加权平均算法(EWA)提出了学习带边信息专家意见的在线投资组合策略(EWAES).然后,从理论上证明了对任何的股票价格序列该策略都能够追踪最优专家意见.最后,采用中美金融市场实际股票数据对EWAES策略进行了数值分析,结果说明了该策略的有效性.

Abstract

Previous online portfolio selection strategies by learning from expert advice often do not employ side information to improve the return of investors.This paper selects the strategies of investing in the same single stock under the same side information state and possibly investing in different single stocks under different side information states as the expert advice.Meanwhile,this paper proposes an online portfolio selection strategy by learning from expert advice with side information(EWAES)based on the exponentially weighted average algorithm(EWA).It is theoretically proved that the strategy can track the optimal expert advice for any stock price sequence.Finally,the numerical analysis for the EWAES strategy using the actual stock data of Chinese and American financial markets shows the effectiveness of this strategy.

关键词

在线投资组合/边信息/专家意见/指数加权平均算法

Key words

online portfolio/side information/expert advice/exponentially weighted average algorithm

引用本文复制引用

基金项目

国家自然科学基金(72371080)

广东省基础与应用基础研究基金(2023A1515012840)

广东省哲学社会科学规划项目(GD23XGL022)

出版年

2024
系统工程学报
中国系统工程学会

系统工程学报

CSTPCDCSCD北大核心
影响因子:1.192
ISSN:1000-5781
参考文献量25
段落导航相关论文