RT Journal Article
SR Electronic
T1 Practical Applications of Deep Reinforcement Learning for Option Replication and Hedging
JF Practical Applications
FD Institutional Investor Journals
SP 1
OP 8
DO 10.3905/pa.9.1.436
VO 9
IS 1
A1 Du, Jiayi (Nicole)
A1 Jin, Muyang
A1 Kolm, Petter N.
A1 Ritter, Gordon
A1 Wang, Yixuan
A1 Zhang, Bofei
YR 2021
UL http://pa.pm-research.com/content/9/1/1.10.abstract
AB If a computer model using machine learning and artificial intelligence like Google’s DeepMind can beat the world’s best human player of the ancient Chinese game of “Go,” can a similar approach help solve the challenge of hedging and replicating option portfolios? Deep Reinforcement Learning for Option Replication and Hedging, from the Fall 2020 issue of The Journal of Financial Data Science, takes a “deep” dive into reinforcement learning approaches for option portfolios. It explores ways of training a computer “agent” by a clever form of trial and error to learn to replicate and hedge an option portfolio directly from data. It’s a positive step forward in the use of modern mathematical approaches to portfolio hedging, and the authors are extending their research in deep reinforcement learning (DRL) to real-world problems in finance, including trading, portfolio management, and hedging.