Full metadata record
DC FieldValueLanguage
dc.contributorDepartment of Computingen_US
dc.contributor.advisorGuo, Song (COMP)en_US
dc.creatorWang, Qi-
dc.identifier.urihttps://theses.lib.polyu.edu.hk/handle/200/11376-
dc.languageEnglishen_US
dc.publisherHong Kong Polytechnic Universityen_US
dc.rightsAll rights reserveden_US
dc.titleTrend adaptive deep reinforcement learning for quantitative tradingen_US
dcterms.abstractQuantitative trading is a crucial investment method in the field of financial investment. Traditional quantitative trading methods usually analyze the price and volume on handcrafted technical indicators. With the rapid development of machine learning and deep learning, quantitative trading based on artificial intelligence has received more and more attention from institutional and individual investors. Reinforcement learning learns through repeated trial and error and can be used to build automated trading strategies without explicit predictions about market trends. Existing works usually use market data and technical indicators of a single stock or futures as environment for reinforcement learning training. However, these methods has several shortcomings which seriously affects the performance such like weak generalization performance and overfitting problem. This paper presents a new trend adaptive reinforcement learning model based on a generative environment. The environment provides real historical market data and simulated market data generated by the artificial neural network as the model's state to improve the generalization performance of the model. At the same time, Trend adaptive agents based on Dueling DQN are proposed in this work. and can have a relatively stable effect for different types of market conditions. The experimental results show that compared with the previous work, the algorithm in this paper effectively increases the profit and reduces the maximum drawdown.en_US
dcterms.extentviii, 41 pages : color illustrationsen_US
dcterms.isPartOfPolyU Electronic Thesesen_US
dcterms.issued2021en_US
dcterms.educationalLevelM.Sc.en_US
dcterms.educationalLevelAll Masteren_US
dcterms.LCSHMachine learningen_US
dcterms.LCSHFinance -- Data processingen_US
dcterms.LCSHHong Kong Polytechnic University -- Dissertationsen_US
dcterms.accessRightsrestricted accessen_US

Files in This Item:
File Description SizeFormat 
5862.pdfFor All Users (off-campus access for PolyU Staff & Students only)1.4 MBAdobe PDFView/Open


Copyright Undertaking

As a bona fide Library user, I declare that:

  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

Show simple item record

Please use this identifier to cite or link to this item: https://theses.lib.polyu.edu.hk/handle/200/11376