No-Transaction Band Action Parameterization in Deep Reinforcement Learning for Option Hedging under Transaction Costs
Resumo
Este trabalho investiga o efeito da parametrização da ação em políticas de aprendizado por reforço para hedge de opções com custos de transação. Comparamos duas formulações com Soft Actor-Critic (SAC): incrementos diretos da posição e bandas de não transação. Com cinco seeds independentes e conjuntos separados de validação e teste, o SAC-Band reduz risco de cauda, custos, frequência de negociação e variabilidade entre treinamentos em relação ao SAC-Trade. Estratégias simples de hedge permanecem competitivas, sugerindo que o principal ganho vem da estrutura financeira incorporada à ação.Referências
Black, F. (1976). The pricing of commodity contracts. Journal of Financial Economics, 3(1–2):167–179.
Black, F. and Scholes, M. (1973). The pricing of options and corporate liabilities. Journal of Political Economy, 81(3):637–654.
Buehler, H., Gonon, L., Teichmann, J., and Wood, B. (2019). Deep hedging. Quantitative Finance, 19(8):1271–1291.
da Silva, A. J., Baczynski, J., and Vicente, J. V. M. (2024). Efficient pricing of path-dependent interest rate derivatives. Applied Stochastic Models in Business and Industry, 40(4):1105–1124.
François, P., Gauthier, G., Godin, F., and Pérez-Mendoza, C. O. (2025). Deep hedging with options using the implied volatility surface. arXiv preprint arXiv:2504.06208.
Haarnoja, T., Zhou, A., Abbeel, P., and Levine, S. (2018). Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor. In Proceedings of the 35th International Conference on Machine Learning, volume 80 of Proceedings of Machine Learning Research, pages 1861–1870. PMLR.
Halperin, I. (2020). Qlbs: Q-learner in the black-scholes(-merton) worlds. The Journal of Derivatives, 28(1):99–122.
Imaki, S., Imajo, K., Ito, K., Minami, K., and Nakagawa, K. (2023). No-transaction band network: A neural network architecture for efficient deep hedging. The Journal of Financial Data Science, 5(2):84–99.
Kolm, P. N. and Ritter, G. (2019). Dynamic replication and hedging: A reinforcement learning approach. The Journal of Financial Data Science, 1(1):159–171.
Ruf, J. and Wang, W. (2020). Neural networks for option pricing and hedging: A literature review. Journal of Computational Finance, 24(1):1–46.
Whalley, A. E. and Wilmott, P. (1997). An asymptotic analysis of an optimal hedging model for option pricing with transaction costs. Mathematical Finance, 7(3):307–324.
Black, F. and Scholes, M. (1973). The pricing of options and corporate liabilities. Journal of Political Economy, 81(3):637–654.
Buehler, H., Gonon, L., Teichmann, J., and Wood, B. (2019). Deep hedging. Quantitative Finance, 19(8):1271–1291.
da Silva, A. J., Baczynski, J., and Vicente, J. V. M. (2024). Efficient pricing of path-dependent interest rate derivatives. Applied Stochastic Models in Business and Industry, 40(4):1105–1124.
François, P., Gauthier, G., Godin, F., and Pérez-Mendoza, C. O. (2025). Deep hedging with options using the implied volatility surface. arXiv preprint arXiv:2504.06208.
Haarnoja, T., Zhou, A., Abbeel, P., and Levine, S. (2018). Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor. In Proceedings of the 35th International Conference on Machine Learning, volume 80 of Proceedings of Machine Learning Research, pages 1861–1870. PMLR.
Halperin, I. (2020). Qlbs: Q-learner in the black-scholes(-merton) worlds. The Journal of Derivatives, 28(1):99–122.
Imaki, S., Imajo, K., Ito, K., Minami, K., and Nakagawa, K. (2023). No-transaction band network: A neural network architecture for efficient deep hedging. The Journal of Financial Data Science, 5(2):84–99.
Kolm, P. N. and Ritter, G. (2019). Dynamic replication and hedging: A reinforcement learning approach. The Journal of Financial Data Science, 1(1):159–171.
Ruf, J. and Wang, W. (2020). Neural networks for option pricing and hedging: A literature review. Journal of Computational Finance, 24(1):1–46.
Whalley, A. E. and Wilmott, P. (1997). An asymptotic analysis of an optimal hedging model for option pricing with transaction costs. Mathematical Finance, 7(3):307–324.
Publicado
19/10/2026
Como Citar
BRAGANÇA, João Felipe da Silva; SILVA, Allan Jonathan da; SANTOS, Fábio Pereira dos.
No-Transaction Band Action Parameterization in Deep Reinforcement Learning for Option Hedging under Transaction Costs. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 23. , 2026, Cuiabá/MT.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 3665-3676.
ISSN 2763-9061.
DOI: https://doi.org/10.5753/eniac.2026.31528.
