An unsupervised learning framework for marketneutral portfolio

作者：

Highlights：

• We show a portfolio optimization based on a novel scheme in assets clustering.

• Significant features of financial assets were explored with data analysis tools.

• A pure alpha strategy for a specific market was explored with machine learning.

• We explore the optimization strategy on several and heterogeneous markets.

摘要

•We show a portfolio optimization based on a novel scheme in assets clustering.•Significant features of financial assets were explored with data analysis tools.•A pure alpha strategy for a specific market was explored with machine learning.•We explore the optimization strategy on several and heterogeneous markets.

论文关键词：Portfolio management,Arbitrage pricing theory,Time series,Cluster analysis

论文评审过程：Received 26 April 2021, Revised 17 November 2021, Accepted 25 November 2021, Available online 20 December 2021, Version of Record 21 December 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116308