Construction of stock portfolios based on k-means clustering of continuous trend features
作者：
Highlights：
• We introduced distance information to cluster stocks based on the continuous trend characteristics of stocks.
• The calculation of returns was revised based on the continuous trend characteristics of stocks.
• The differences in the results of RT and RC were compared by statistical test.
• A classification model was trained to avoid the decline risk of portfolio.
• Results were significantly improved, and the optimal number of stocks in the portfolio could be determined.
摘要
•We introduced distance information to cluster stocks based on the continuous trend characteristics of stocks.•The calculation of returns was revised based on the continuous trend characteristics of stocks.•The differences in the results of RT and RC were compared by statistical test.•A classification model was trained to avoid the decline risk of portfolio.•Results were significantly improved, and the optimal number of stocks in the portfolio could be determined.
论文关键词：Stock portfolio,k-means clustering,Sharpe ratio,Continuous trend characteristics,Portfolio theory
论文评审过程：Received 4 December 2021, Revised 29 June 2022, Accepted 29 June 2022, Available online 14 July 2022, Version of Record 19 July 2022.
论文官网地址：https://doi.org/10.1016/j.knosys.2022.109358
