Gamma and vega hedging using deep distributional reinforcement learning
We show how reinforcement learning can be used in conjunction with quantile regression to develop a hedging strategy for a trader responsible for derivatives that arrive stochastically and depend on a single underlying asset. We assume that the trader makes the portfolio delta-neutral at the end of...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-02-01
|
| Series: | Frontiers in Artificial Intelligence |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/frai.2023.1129370/full |
| _version_ | 1797900458825613312 |
|---|---|
| author | Jay Cao Jacky Chen Soroush Farghadani John Hull Zissis Poulos Zeyu Wang Jun Yuan |
| author_facet | Jay Cao Jacky Chen Soroush Farghadani John Hull Zissis Poulos Zeyu Wang Jun Yuan |
| author_sort | Jay Cao |
| collection | DOAJ |
| description | We show how reinforcement learning can be used in conjunction with quantile regression to develop a hedging strategy for a trader responsible for derivatives that arrive stochastically and depend on a single underlying asset. We assume that the trader makes the portfolio delta-neutral at the end of each day by taking a position in the underlying asset. We focus on how trades in options can be used to manage gamma and vega. The option trades are subject to transaction costs. We consider three different objective functions. We reach conclusions on how the optimal hedging strategy depends on the trader's objective function, the level of transaction costs, and the maturity of the options used for hedging. We also investigate the robustness of the hedging strategy to the process assumed for the underlying asset. |
| first_indexed | 2024-04-10T08:46:21Z |
| format | Article |
| id | doaj.art-e6ec3324d6994697b2efba26803914cb |
| institution | Directory Open Access Journal |
| issn | 2624-8212 |
| language | English |
| last_indexed | 2024-04-10T08:46:21Z |
| publishDate | 2023-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Artificial Intelligence |
| spelling | doaj.art-e6ec3324d6994697b2efba26803914cb2023-02-22T08:24:55ZengFrontiers Media S.A.Frontiers in Artificial Intelligence2624-82122023-02-01610.3389/frai.2023.11293701129370Gamma and vega hedging using deep distributional reinforcement learningJay Cao0Jacky Chen1Soroush Farghadani2John Hull3Zissis Poulos4Zeyu Wang5Jun Yuan6Joseph L. Rotman School of Management, University of Toronto, Toronto, ON, CanadaJoseph L. Rotman School of Management, University of Toronto, Toronto, ON, CanadaDepartment of Computer Science, University of Toronto, Toronto, ON, CanadaJoseph L. Rotman School of Management, University of Toronto, Toronto, ON, CanadaJoseph L. Rotman School of Management, University of Toronto, Toronto, ON, CanadaJoseph L. Rotman School of Management, University of Toronto, Toronto, ON, CanadaJoseph L. Rotman School of Management, University of Toronto, Toronto, ON, CanadaWe show how reinforcement learning can be used in conjunction with quantile regression to develop a hedging strategy for a trader responsible for derivatives that arrive stochastically and depend on a single underlying asset. We assume that the trader makes the portfolio delta-neutral at the end of each day by taking a position in the underlying asset. We focus on how trades in options can be used to manage gamma and vega. The option trades are subject to transaction costs. We consider three different objective functions. We reach conclusions on how the optimal hedging strategy depends on the trader's objective function, the level of transaction costs, and the maturity of the options used for hedging. We also investigate the robustness of the hedging strategy to the process assumed for the underlying asset.https://www.frontiersin.org/articles/10.3389/frai.2023.1129370/fullderivativeshedginggammavegadelta-neutralreinforcement learning |
| spellingShingle | Jay Cao Jacky Chen Soroush Farghadani John Hull Zissis Poulos Zeyu Wang Jun Yuan Gamma and vega hedging using deep distributional reinforcement learning Frontiers in Artificial Intelligence derivatives hedging gamma vega delta-neutral reinforcement learning |
| title | Gamma and vega hedging using deep distributional reinforcement learning |
| title_full | Gamma and vega hedging using deep distributional reinforcement learning |
| title_fullStr | Gamma and vega hedging using deep distributional reinforcement learning |
| title_full_unstemmed | Gamma and vega hedging using deep distributional reinforcement learning |
| title_short | Gamma and vega hedging using deep distributional reinforcement learning |
| title_sort | gamma and vega hedging using deep distributional reinforcement learning |
| topic | derivatives hedging gamma vega delta-neutral reinforcement learning |
| url | https://www.frontiersin.org/articles/10.3389/frai.2023.1129370/full |
| work_keys_str_mv | AT jaycao gammaandvegahedgingusingdeepdistributionalreinforcementlearning AT jackychen gammaandvegahedgingusingdeepdistributionalreinforcementlearning AT soroushfarghadani gammaandvegahedgingusingdeepdistributionalreinforcementlearning AT johnhull gammaandvegahedgingusingdeepdistributionalreinforcementlearning AT zissispoulos gammaandvegahedgingusingdeepdistributionalreinforcementlearning AT zeyuwang gammaandvegahedgingusingdeepdistributionalreinforcementlearning AT junyuan gammaandvegahedgingusingdeepdistributionalreinforcementlearning |