主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2023, Vol. 31 ›› Issue (12): 96-106.doi: 10.16381/j.cnki.issn1003-207x.2021.2308

• • 上一篇    下一篇

基于机器学习预测股票收益率的两步骤M-SV投资组合优化

张鹏1(),党世力1,黄梅雨2,李璟欣1   

  1. 1.华南师范大学经济与管理学院, 广东 广州 510006
    2.华中科技大学管理学院, 湖北 武汉 430074
  • 收稿日期:2021-11-08 修回日期:2022-03-04 出版日期:2023-12-15 发布日期:2024-01-06
  • 通讯作者: 张鹏 E-mail:zhangpeng300478@aliyun.com
  • 基金资助:
    国家自然科学基金资助项目(71271161);广东省社会科学项目(GD19CGL32)

Two-stage Mean Semi-variance Portfolio Optimization with Stock Return Prediction Using Machine Learning

Peng ZHANG1(),Shi-li DANG1,Mei-yu HUANG2,Jing-xin LI1   

  1. 1.School of Economics & Management, South China Normal University, Guangzhou 510006, China
    2.School of Management, Huazhong University of Science & Technology, Wuhan 430074, China
  • Received:2021-11-08 Revised:2022-03-04 Online:2023-12-15 Published:2024-01-06
  • Contact: Peng ZHANG E-mail:zhangpeng300478@aliyun.com

摘要:

由于准确预测股票收益序列能够提高投资组合优化模型的表现,相比于传统的计量经济预测模型,机器学习在处理非线性和非平稳特征的问题上更具优势。因此,本文提出了一种基于机器学习方法的两步骤多元化投资组合优化模型。具体而言,该模型包括以下两个步骤:步骤1是股票选择,即通过机器学习方法极端梯度提升法(extreme gradient boosting,XGBoost)、支持向量回归(support vector regression,SVR)、K近邻算法(K-nearest neighbor,KNN)选择具有较高预测收益率的股票,并对模型进行评估和选择。步骤2是投资组合优化,在考虑交易成本、上下界约束的现实约束条件下,采用均值-下半方差(mean semi-variance, M-SV)模型、均值-方差模型和等比例模型确定所选股票的投资比例。最后,以沪深300指数成分股作为研究样本,实证结果表明,XGBoost+M-SV模型在收益和风险指标上均优于其他模型和沪深300指数。

关键词: 均值-下半方差, 旋转算法, 机器学习, 股票收益预测

Abstract:

Since accurately predicting stock return sequences can improve the performance of portfolio optimization models, the results have indicated that machine learning methods have a greater capacity to confront problems with nonlinear, nonstationary charateristics than econometric models. Consequently, a novel two-stage method is proposed for well-diversified portfolio construction based on stock return prediction using machine learning, which includes two stages. To be specific, the purpose of the first stage is to select diversified stocks with high predicted returns, where the returns are predicted by machine learning methods, i.e. eXtreme Gradient Boosting(XGBoost), support vector regression(SVR), K-Nearest Neighbor(KNN), and evaluate and select the model. In the second stage, considering the constraints such as transaction costs and threshold constraints, the predictive results are incorporated into the mean semi-variance (M-SV) model, mean-variance model and equally weighted model to determine optimal portfolio. Finally, using China Securities 300 Index component stocks as study sample, the empirical results demonstrate that the XGBoost+MSV model achieves better results than similar counterparts and market index in terms of return and return-risk metrics.

Key words: mean semi-variance, pivoting algorithm, machine learning, stock price prediction

中图分类号: