Feature Transformation and Simulation of Short Term Price Variability in Reinforcement Learning for Portfolio Management

Yen Chih Lin, Jeremy Blum

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Reinforcement learning has been shown capable of learning optimal strategies from imperfect information environments in order to create robust decision support systems. This paper shows that two automatic feature transformation techniques - Bayesian recurrent neural network (BRNN) for modelling future price trends and Generative Adversarial Networks (GANs) for modelling short-term realistic price variability - are able to improve the performance of reinforcement learning agents in solving portfolio management problem effectively, when measured in terms of increasing profitability and reducing risks.

Original languageEnglish (US)
Title of host publicationProceedings of the 2020 Spring Simulation Conference, SpringSim 2020
EditorsFernando J. Barros, Xiaolin Hu, Hamdi Kavak, Alberto A. Del Barrio
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781565553705
DOIs
StatePublished - May 2020
Event2020 Spring Simulation Conference, SpringSim 2020 - Virtual, Fairfax, United States
Duration: May 18 2020May 21 2020

Publication series

NameProceedings of the 2020 Spring Simulation Conference, SpringSim 2020

Conference

Conference2020 Spring Simulation Conference, SpringSim 2020
CountryUnited States
CityVirtual, Fairfax
Period5/18/205/21/20

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Computer Vision and Pattern Recognition
  • Information Systems and Management
  • Safety, Risk, Reliability and Quality
  • Modeling and Simulation

Fingerprint Dive into the research topics of 'Feature Transformation and Simulation of Short Term Price Variability in Reinforcement Learning for Portfolio Management'. Together they form a unique fingerprint.

Cite this