RLOP: RL Methods in Option Pricing from a Mathematical Perspective

TL;DR

This paper introduces two mathematically grounded environments for applying reinforcement learning to option pricing, comparing learned hedging strategies with classical models and analyzing various influencing factors.

Contribution

It develops the RLOP and modified QLBS environments for RL in option pricing, integrating neural network-based agents and providing a comparative analysis with Black-Scholes predictions.

Findings

RL methods can effectively learn hedging strategies

Optimal prices are influenced by various market factors

The approach bridges reinforcement learning with mathematical finance

Abstract

Abstract In this work, we build two environments, namely the modified QLBS and RLOP models, from a mathematics perspective which enables RL methods in option pricing through replicating by portfolio. We implement the environment specifications (the source code can be found at https://github.com/owen8877/RLOP), the learning algorithm, and agent parametrization by a neural network. The learned optimal hedging strategy is compared against the BS prediction. The effect of various factors is considered and studied based on how they affect the optimal price and position.