Pricing and hedging American-style options with deep learning

TL;DR

This paper presents a deep learning approach for accurately pricing and hedging American-style options by estimating optimal stopping policies, bounds, and dynamic strategies, demonstrated on Bermudan max-call options.

Contribution

It introduces a novel deep learning framework that computes bounds and hedging strategies for American options, improving accuracy and efficiency over traditional methods.

Findings

Highly accurate option prices and hedging strategies

Small replication errors across tested options

Effective estimation of bounds and stopping policies

Abstract

In this paper we introduce a deep learning method for pricing and hedging American-style options. It first computes a candidate optimal stopping policy. From there it derives a lower bound for the price. Then it calculates an upper bound, a point estimate and confidence intervals. Finally, it constructs an approximate dynamic hedging strategy. We test the approach on different specifications of a Bermudan max-call option. In all cases it produces highly accurate prices and dynamic hedging strategies with small replication errors.