Deep Hedging: Learning to Simulate Equity Option Markets

13 Pages Posted: 14 Nov 2019

Date Written: October 16, 2019

Abstract

We construct realistic equity option market simulators based on generative adversarial networks (GANs). We consider recurrent and temporal convolutional architectures, and assess the impact of state compression. Option market simulators are highly relevant because they allow us to extend the limited real-world data sets available for the training and evaluation of option trading strategies. We show that network-based generators outperform classical methods on a range of benchmark metrics, and adversarial training achieves the best performance. Our work demonstrates for the first time that GANs can be successfully applied to the task of generating multivariate financial time series.

Keywords: volatility surface, generative modeling, generative adversarial networks, mathematical finance, time series, neural networks, options

JEL Classification: C15, C45, C5, C53, C6, C63, G00

Suggested Citation

Wiese, Magnus and Bai, Lianjun and Wood, Ben and Buehler, Hans, Deep Hedging: Learning to Simulate Equity Option Markets (October 16, 2019). Available at SSRN: https://ssrn.com/abstract=3470756 or http://dx.doi.org/10.2139/ssrn.3470756

Magnus Wiese

JP Morgan ( email )

London
United Kingdom

Lianjun Bai

JP Morgan ( email )

London
United Kingdom

Ben Wood (Contact Author)

JP Morgan Chase ( email )

London
United Kingdom

Hans Buehler

JP Morgan ( email )

London
United Kingdom

Register to save articles to
your library

Register

Paper statistics

Downloads
382
Abstract Views
1,224
rank
81,154
PlumX Metrics