Abstract:With the rapid advancement of neural networks, methods for option pricing have evolved significantly. This study employs the Black-Scholes-Merton (B-S-M) model, incorporating an additional variable to improve the accuracy of predictions compared to the traditional Black-Scholes (B-S) model. Furthermore, Convolutional Kolmogorov-Arnold Networks (Conv-KANs) and Kolmogorov-Arnold Networks (KANs) are introduced to demonstrate that networks with enhanced non-linear capabilities yield superior fitting performance. For comparative analysis, Conv-LSTM and LSTM models, which are widely used in time series forecasting, are also applied. Additionally, a novel data selection strategy is proposed to simulate a real trading environment, thereby enhancing the robustness of the model.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to improve the accuracy of option pricing. Specifically, the authors introduce Convolutional Kolmogorov - Arnold Networks (Conv - KANs) and Kolmogorov - Arnold Networks (KANs) to enhance the non - linear fitting ability and improve the traditional Black - Scholes (B - S) model and Black - Scholes - Merton (B - S - M) model. ### Main problems 1. **Improve the accuracy of option pricing**: - The traditional B - S and B - S - M models have deviations and limitations in practical applications, especially when dealing with dividend payments. - Introduce new neural network models (such as Conv - KANs and KANs) to improve the prediction accuracy. 2. **Simulate the real trading environment**: - Propose a new data selection strategy. By adding white noise to simulate the real financial market environment, thus enhancing the robustness and generalization ability of the model. 3. **Compare the performance of different models**: - Use Conv - LSTM and LSTM models for time - series prediction, and conduct a comparative analysis with the newly proposed Conv - KANs and KANs models to verify the effectiveness of the new models. ### Key methods and techniques - **Black - Scholes - Merton (B - S - M) model**: It improves the classic B - S model by adding the dividend rate \( q \) to more accurately reflect the real - market situation. \[ c = S_0 e^{-qT} N(d_1) - K e^{-rT} N(d_2) \] \[ p = K e^{-rT} N(-d_2) - S_0 e^{-qT} N(-d_1) \] where, \[ d_1 = \frac{\ln(S_0 / K) + (r - q + \sigma^2 / 2) T}{\sigma \sqrt{T}} \] \[ d_2 = d_1 - \sigma \sqrt{T} \] - **Convolutional Kolmogorov - Arnold Networks (Conv - KANs)**: Combine the convolution operation and the non - linear fitting ability of KANs to capture the complex patterns in option prices. - **Data processing method**: In order to better simulate the real trading environment, a certain amount of noise is retained, and the data is divided into training sets and test sets to ensure that the model has strong generalization ability. Through these methods and techniques, the paper aims to provide more accurate and reliable tools for financial derivatives pricing.

Option Pricing with Convolutional Kolmogorov-Arnold Networks

KANOP: A Data-Efficient Option Pricing Model using Kolmogorov-Arnold Networks

MLP, XGBoost, KAN, TDNN, and LSTM-GRU Hybrid RNN with Attention for SPX and NDX European Call Option Pricing

Improving option price forecasts with neural networks and support vector regressions

Option Pricing by Neural Stochastic Differential Equations: A Simulation-Optimization Approach

A new Input Convex Neural Network with application to options pricing

Pricing options with a new hybrid neural network model

Machine learning for option pricing: an empirical investigation of network architectures

American Call Options Pricing With Modular Neural Networks

A Simple Method of Forecasting Option Prices Based on Neural Networks

Deep Calibration With Artificial Neural Network: A Performance Comparison on Option Pricing Models

Financial option valuation by unsupervised learning with artificial neural networks

KAN based Autoencoders for Factor Models

Pricing Options and Computing Implied Volatilities using Neural Networks

The Study on Option Pricing Based on Wiener–Itô Chaos Expansion and Generative Adversarial Networks

Jump Diffusion-Informed Neural Networks with Transfer Learning for Accurate American Option Pricing under Data Scarcity

Kolmogorov-Arnold Networks (KAN) for Time Series Classification and Robust Analysis

KAN: Kolmogorov-Arnold Networks

Convolutional Kolmogorov-Arnold Networks

Kolmogorov-Arnold Networks for Time Series: Bridging Predictive Power and Interpretability

Neural Network Learning of Black-Scholes Equation for Option Pricing