Machine learning classification and regression models for predicting directional changes trend reversal in FX markets
作者：
Highlights：
• A novel trading strategy based on the event-based concept of directional changes.
• The trading strategy includes classification and regression algorithms.
• Algorithm tested on 1000 datasets from 20 FX currency pairs.
• Proposed approach is able to generate new and profitable trading strategies.
• Proposed approach significantly outperforms all other benchmarks.
摘要
•A novel trading strategy based on the event-based concept of directional changes.•The trading strategy includes classification and regression algorithms.•Algorithm tested on 1000 datasets from 20 FX currency pairs.•Proposed approach is able to generate new and profitable trading strategies.•Proposed approach significantly outperforms all other benchmarks.
论文关键词：Directional changes,Regression,Classification,Genetic programming,Forex/FX,Machine learning
论文评审过程：Received 3 August 2020, Revised 19 January 2021, Accepted 19 January 2021, Available online 29 January 2021, Version of Record 19 February 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.114645
