Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A Machine Learning based Pairs Trading Investment StrategyProposed Pairs Selection Framework

A Machine Learning based Pairs Trading Investment Strategy: Proposed Pairs Selection Framework [This chapter proposes a new framework for pairs selection aiming to answer this work’s first research question (introduced in Chap. 1): “Can Unsupervised Learning find more promising pairs?”. It starts by describing the problems associated with the most commonly used approaches to select pairs, and how they may guide us in the quest for a novel approach. Next, a framework composed of 3 stages, (i) dimensionality reduction, (ii) Unsupervised Learning and (iii) pairs selection, is introduced. Each stage is described separately. First, a useful tool for reducing data dimensionality, Principal Component Analysis, is explored. Then, the pursuit of the most suitable Unsupervised Learning algorithm for clustering is followed in detail. Finally, the criteria adopted to select potentially profitable pairs from the clusters previously formed are outlined. In the end, a summary diagram connecting the different concepts introduced throughout the chapter is presented, for consolidation purposes.] http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

A Machine Learning based Pairs Trading Investment StrategyProposed Pairs Selection Framework

Download PDF
14 pages

Loading next page...
 
/lp/springer-journals/a-machine-learning-based-pairs-trading-investment-strategy-proposed-yl49t0VVsf
Publisher
Springer International Publishing
Copyright
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021
ISBN
978-3-030-47250-4
Pages
21 –35
DOI
10.1007/978-3-030-47251-1_3
Publisher site
See Chapter on Publisher Site

Abstract

[This chapter proposes a new framework for pairs selection aiming to answer this work’s first research question (introduced in Chap. 1): “Can Unsupervised Learning find more promising pairs?”. It starts by describing the problems associated with the most commonly used approaches to select pairs, and how they may guide us in the quest for a novel approach. Next, a framework composed of 3 stages, (i) dimensionality reduction, (ii) Unsupervised Learning and (iii) pairs selection, is introduced. Each stage is described separately. First, a useful tool for reducing data dimensionality, Principal Component Analysis, is explored. Then, the pursuit of the most suitable Unsupervised Learning algorithm for clustering is followed in detail. Finally, the criteria adopted to select potentially profitable pairs from the clusters previously formed are outlined. In the end, a summary diagram connecting the different concepts introduced throughout the chapter is presented, for consolidation purposes.]

Published: Jul 14, 2020

Keywords: Dimensionality reduction; PCA; Unsupervised Learning; DBSCAN; OPTICS

There are no references for this article.