Deep Partial Hedging

TL;DR

This paper demonstrates that deep learning can effectively train neural networks to perform partial hedging, incorporating realistic market features like transaction costs and discrete trading intervals.

Contribution

It introduces a neural network-based approach to partial hedging that improves realism and flexibility over traditional methods.

Findings

Neural networks successfully replicate partial hedging payoffs.

The approach accommodates transaction costs and market dynamics.

Enhanced hedging performance in practical trading scenarios.

Abstract

Using techniques from deep learning (cf. [B\"uh+19]), we show that neural networks can be trained successfully to replicate the modified payoff functions that were first derived in the context of partial hedging by [FL00]. Not only does this approach better accommodate the realistic setting of hedging in discrete time, it also allows for the inclusion of transaction costs as well as general market dynamics.