Machine learning in pair trading

This research seeks to develop and compare traditional threshold methods and machine learning models applied in pair trading of stocks. Unsupervised machine learning firstly segments stocks into different clusters based on stocks’ return and volatility profile. Stock pairs would be filtered ou...

Full description

Saved in:
Bibliographic Details
Main Author: Wang, Guanlan
Other Authors: Bo An
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2022
Subjects:
Online Access:https://hdl.handle.net/10356/156581
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:This research seeks to develop and compare traditional threshold methods and machine learning models applied in pair trading of stocks. Unsupervised machine learning firstly segments stocks into different clusters based on stocks’ return and volatility profile. Stock pairs would be filtered out according to several criteria. Traditional threshold method, XGBoost, and LSTM models are applied to the pair trading mechanism with performance compared.