Deep graph convolutional reinforcement learning for financial portfolio management – DeepPocket

•Portfolio management using a deep graph convolutional reinforcement learning method.•Extracting low-dimensional features using Restricted Stacked Autoencoder.•Interrelation among financial instruments is obtained using a DeepPocket method.•An actor-critic framework is exploited to enforce the inves...

Full description

Saved in:
Bibliographic Details
Published in:Expert systems with applications Vol. 182; p. 115127
Main Authors: Soleymani, Farzan, Paquet, Eric
Format: Journal Article
Language:English
Published: New York Elsevier Ltd 15.11.2021
Elsevier
Elsevier BV
Subjects:
Actor-critic
Deep reinforcement learning
Feature extraction
Financial instruments
Graph convolutional network
Graph theory
Graphical representations
Investment strategy
Learning
Online leaning
online learning
Optimization
Portfolio management
Restricted stacked autoencoder
Return on investment
Restricted stacked autoencoder
Deep reinforcement learning
Online leaning
Actor-critic
Graph convolutional network
Portfolio management
ISSN:0957-4174, 1873-6793
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Portfolio management using a deep graph convolutional reinforcement learning method.•Extracting low-dimensional features using Restricted Stacked Autoencoder.•Interrelation among financial instruments is obtained using a DeepPocket method.•An actor-critic framework is exploited to enforce the investment policy.•The reinforcement learning framework is trained both offline and online. Portfolio management aims at maximizing the return on investment while minimizing risk by continuously reallocating the assets forming the portfolio. These assets are not independent but correlated during a short time period. A graph convolutional reinforcement learning framework called DeepPocket is proposed whose objective is to exploit the time-varying interrelations between financial instruments. These interrelations are represented by a graph whose nodes correspond to the financial instruments while the edges correspond to a pair-wise correlation function in between assets. DeepPocket consists of a restricted, stacked autoencoder for feature extraction, a convolutional network to collect underlying local information shared among financial instruments and an actor-critic reinforcement learning agent. The actor-critic structure contains two convolutional networks in which the actor learns and enforces an investment policy which is, in turn, evaluated by the critic in order to determine the best course of action by constantly reallocating the various portfolio assets to optimize the expected return on investment. The agent is initially trained offline with online stochastic batching on historical data. As new data become available, it is trained online with a passive concept drift approach to handle unexpected changes in their distributions. DeepPocket is evaluated against five real-life datasets over three distinct investment periods, including during the Covid-19 crisis, and clearly outperformed market indexes.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0957-4174
1873-6793
DOI:10.1016/j.eswa.2021.115127