Optimal Option Hedging with Policy Gradient

Bo Xiao; Wuguannan Yao; Xiang Zhou

doi:10.1109/ICDMW53433.2021.00145

Optimal Option Hedging with Policy Gradient

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

3 Citations (Scopus)

Abstract

For the option hedging problem, we propose a reinforcement learning framework to statistically hedge generic financial options under the setting of non-zero risk-free interest rate and transaction cost. Instead of minimizing the risk of final wealth, we maximize the cumulative reward during the whole hedging process to find the optimal hedging strategies. In particular, we define the fundamental elements of reinforcement learning for the hedging problem. To illustrate the formulation, we utilize the Monte Carlo policy gradient with baseline onto both synthetic and realistic data to learn the state-value function and optimal policy function. Empirical evaluations demonstrate the effectiveness and performance of the proposed approach.

Original language	English
Title of host publication	Proceedings - 21st IEEE International Conference on Data Mining Workshops, ICDMW 2021
Editors	Bing Xue, Mykola Pechenizkiy, Yun Sing Koh
Publisher	IEEE Computer Society
Pages	1112-1119
ISBN (Electronic)	9781665424271
ISBN (Print)	9781665424288
DOIs	https://doi.org/10.1109/ICDMW53433.2021.00145
Publication status	Published - Dec 2021
Event	21st IEEE International Conference on Data Mining (ICDM 2021) - Virtual, Auckland, New Zealand Duration: 7 Dec 2021 → 10 Dec 2021 https://icdm2021.auckland.ac.nz/

Publication series

Name	IEEE International Conference on Data Mining Workshops, ICDMW
Volume	2021-December
ISSN (Print)	2375-9232
ISSN (Electronic)	2375-9259

Conference

Conference	21st IEEE International Conference on Data Mining (ICDM 2021)
Abbreviated title	IEEE ICDM 2021
Place	New Zealand
City	Auckland
Period	7/12/21 → 10/12/21
Internet address	https://icdm2021.auckland.ac.nz/

Research Keywords

Option Hedging
Policy Gradient
Reinforcement Learning

RGC Funding Information

RGC-funded

Access Output

10.1109/ICDMW53433.2021.00145

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Xiao, B., Yao, W., & Zhou, X. (2021). Optimal Option Hedging with Policy Gradient. In B. Xue, M. Pechenizkiy, & Y. S. Koh (Eds.), Proceedings - 21st IEEE International Conference on Data Mining Workshops, ICDMW 2021 (pp. 1112-1119). (IEEE International Conference on Data Mining Workshops, ICDMW; Vol. 2021-December). IEEE Computer Society. https://doi.org/10.1109/ICDMW53433.2021.00145

@inproceedings{310dffacbdca4058b8ac3d10a31c036a,

title = "Optimal Option Hedging with Policy Gradient",

abstract = "For the option hedging problem, we propose a reinforcement learning framework to statistically hedge generic financial options under the setting of non-zero risk-free interest rate and transaction cost. Instead of minimizing the risk of final wealth, we maximize the cumulative reward during the whole hedging process to find the optimal hedging strategies. In particular, we define the fundamental elements of reinforcement learning for the hedging problem. To illustrate the formulation, we utilize the Monte Carlo policy gradient with baseline onto both synthetic and realistic data to learn the state-value function and optimal policy function. Empirical evaluations demonstrate the effectiveness and performance of the proposed approach.",

keywords = "Option Hedging, Policy Gradient, Reinforcement Learning",

author = "Bo Xiao and Wuguannan Yao and Xiang Zhou",

year = "2021",

month = dec,

doi = "10.1109/ICDMW53433.2021.00145",

language = "English",

isbn = "9781665424288",

series = "IEEE International Conference on Data Mining Workshops, ICDMW",

publisher = "IEEE Computer Society",

pages = "1112--1119",

editor = "Bing Xue and Mykola Pechenizkiy and Koh, {Yun Sing}",

booktitle = "Proceedings - 21st IEEE International Conference on Data Mining Workshops, ICDMW 2021",

address = "United States",

note = "21st IEEE International Conference on Data Mining (ICDM 2021), IEEE ICDM 2021 ; Conference date: 07-12-2021 Through 10-12-2021",

url = "https://icdm2021.auckland.ac.nz/",

}

Xiao, B , Yao, W & Zhou, X 2021, Optimal Option Hedging with Policy Gradient. in B Xue, M Pechenizkiy & YS Koh (eds), Proceedings - 21st IEEE International Conference on Data Mining Workshops, ICDMW 2021. IEEE International Conference on Data Mining Workshops, ICDMW, vol. 2021-December, IEEE Computer Society, pp. 1112-1119, 21st IEEE International Conference on Data Mining (ICDM 2021), Auckland, New Zealand, 7/12/21. https://doi.org/10.1109/ICDMW53433.2021.00145

Optimal Option Hedging with Policy Gradient. / Xiao, Bo ; Yao, Wuguannan ; Zhou, Xiang.
Proceedings - 21st IEEE International Conference on Data Mining Workshops, ICDMW 2021. ed. / Bing Xue; Mykola Pechenizkiy; Yun Sing Koh. IEEE Computer Society, 2021. p. 1112-1119 (IEEE International Conference on Data Mining Workshops, ICDMW; Vol. 2021-December).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Optimal Option Hedging with Policy Gradient

AU - Xiao, Bo

AU - Yao, Wuguannan

AU - Zhou, Xiang

PY - 2021/12

Y1 - 2021/12

N2 - For the option hedging problem, we propose a reinforcement learning framework to statistically hedge generic financial options under the setting of non-zero risk-free interest rate and transaction cost. Instead of minimizing the risk of final wealth, we maximize the cumulative reward during the whole hedging process to find the optimal hedging strategies. In particular, we define the fundamental elements of reinforcement learning for the hedging problem. To illustrate the formulation, we utilize the Monte Carlo policy gradient with baseline onto both synthetic and realistic data to learn the state-value function and optimal policy function. Empirical evaluations demonstrate the effectiveness and performance of the proposed approach.

AB - For the option hedging problem, we propose a reinforcement learning framework to statistically hedge generic financial options under the setting of non-zero risk-free interest rate and transaction cost. Instead of minimizing the risk of final wealth, we maximize the cumulative reward during the whole hedging process to find the optimal hedging strategies. In particular, we define the fundamental elements of reinforcement learning for the hedging problem. To illustrate the formulation, we utilize the Monte Carlo policy gradient with baseline onto both synthetic and realistic data to learn the state-value function and optimal policy function. Empirical evaluations demonstrate the effectiveness and performance of the proposed approach.

KW - Option Hedging

KW - Policy Gradient

KW - Reinforcement Learning

UR - http://www.scopus.com/inward/record.url?scp=85125322263&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85125322263&origin=recordpage

U2 - 10.1109/ICDMW53433.2021.00145

DO - 10.1109/ICDMW53433.2021.00145

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9781665424288

T3 - IEEE International Conference on Data Mining Workshops, ICDMW

SP - 1112

EP - 1119

BT - Proceedings - 21st IEEE International Conference on Data Mining Workshops, ICDMW 2021

A2 - Xue, Bing

A2 - Pechenizkiy, Mykola

A2 - Koh, Yun Sing

PB - IEEE Computer Society

T2 - 21st IEEE International Conference on Data Mining (ICDM 2021)

Y2 - 7 December 2021 through 10 December 2021

ER -

Optimal Option Hedging with Policy Gradient

Abstract

Publication series

Conference

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this