Studies of Issues on Hybrid Neural Networks for Pricing Financial Options

Abstract

Neural networks (NN) in combination with parametric models (i.e., Hybrid models) are increasingly employed for option pricing. However, a fundamental question needs to be addressed within the current research domain of financial option pricing utilizing hybrid NN: Does integrating a NN with a more advanced mathematical option model enhance its pricing capabilities compared to integrating with a robust mathematical model? In this paper, we conducted a novel ANN-Heston and ANN-CS option pricing research based on the 50ETF options obtained from the Shanghai Stock Exchange covering January 2018 to December 2021. Having compared the pricing accuracies between ANN-Heston and ANN-CS, we show that the hybrid ANN in combination with the CS model is adequately competent in pricing Chinese options. We also comment that the parametric model should be robust with only some parameters to be estimated. The CS model can capture Chinese option features, and its hybrid ANN model exhibits remarkable competence in pricing options. This research is useful for practitioners and researchers in the field of option trading.

Received: 7 May 2024 | Revised: 23 October 2024 | Accepted: 10 November 2024

Conflicts of Interest

The author declares that he has no conflicts of interest to this work.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Author Contribution Statement

David Liu: Conceptualization, Methodology, Formal analysis, Resources, Writing – original draft, Writing – review & editing, Visualization, Validation, Investigation.