Journal of Computational Finance
ISSN:
1460-1559 (print)
1755-2850 (online)
Editor-in-chief: Christoph Reisinger
Volume 27, Number 2 (September 2023)
Editor's Letter
Christoph Reisinger
University of Oxford
It is a great pleasure to introduce this issue of The Journal of Computational Finance, which includes such topics as the refinement of classical techniques for volatility surfaces, modern adjoint techniques for model calibration, and the continuing development of hedging strategies using state-of-the-art reinforcement learning.
In the issue’s first paper, “Refined analysis of the no-butterfly-arbitrage domain for SSVI slices”, Claude Martini and Arianna Mingone provide a detailed analysis of the inputs to the popular surface stochastic-volatility-inspired (SSVI) parametric volatility surface model such that standard arbitrage conditions are satisfied. Various simplifications are made to simplify calibration procedures.
Next, in the second paper in the issue, José Brito, Andrei Goloubentsev and Evgeny Goncharov give us “Automatic adjoint differentiation for special functions involving expectations”. Their methods are targeted at an important class of objectives that naturally arise from parametric model calibration with Monte Carlo methods. Using adapted formulations of algorithmic adjoint differentiation in tandem with parallelization, the new algorithms are faster and easier to implement than previous methods.
There is a growing body of literature documenting recent advances in both industry and academia regarding market scenario generation and data-driven hedging. In our final paper, “Hedging of financial derivative contracts via Monte Carlo tree search”, Oleg Szehr presents a specific approach that builds on an AlphaZero version of Monte Carlo tree search. Using only contractual cashflows as inputs to utility maximization, his proposed method is shown to perform on a par with, or be superior to, contemporary Q-learning and deep-hedging methods on basic hedging tasks.
Finally, let me draw your attention to the International Conference on Computational Finance 2024 (ICCF24), whose themes align strongly with this journal. We expect that many of our authors and editorial board members will present at this conference, which will be held in Amsterdam on April 2–5, 2024, and we look forward to devoting a special issue to the conference papers in a subsequent volume. The conference’s main organizer is my predecessor as editor-in-chief, Cornelis W. (“Kees”) Oosterlee, and together we will be hosting a special session to celebrate the many contributions to the field by Peter Forsyth, another former editor of The Journal of Computational Finance, on the occasion of his seventieth birthday. The recipient of the 2022 Journal of Computational Finance Young Researcher Award (the Peter Carr Memorial Prize), Álvaro Leitao Rodríguez, will deliver a plenary lecture, and there will be talks from renowned and upcoming academics and practitioners in the field of quantitative finance. I am greatly looking forward to this exciting conference and hope to see many of you there. In the meantime, I hope you find this issue of The Journal of Computational Finance interesting.
Papers in this issue
Refined analysis of the no-butterfly-arbitrage domain for SSVI slices
The authors investigate the surface SVI model with three with three parameters, applying the SVI results to give the nobutterfly- arbitrage domain
Automatic adjoint differentiation for special functions involving expectations
The authors put forward AAD algorithms for functions involving expectations and use their technique to calibrate European options.
Hedging of financial derivative contracts via Monte Carlo tree search
This paper applies the Monte Carlo tree search as a method for replication in the presence of risk and market friction