American Option Pricing using Self-Attention GRU and Shapley Value Interpretation

Yanhui Shen

2023-10-19

Abstract:Options, serving as a crucial financial instrument, are used by investors to manage and mitigate their investment risks within the securities market. Precisely predicting the present price of an option enables investors to make informed and efficient decisions. In this paper, we propose a machine learning method for forecasting the prices of SPY (ETF) option based on gated recurrent unit (GRU) and self-attention mechanism. We first partitioned the raw dataset into 15 subsets according to moneyness and days to maturity criteria. For each subset, we matched the corresponding U.S. government bond rates and Implied Volatility Indices. This segmentation allows for a more insightful exploration of the impacts of risk-free rates and underlying volatility on option pricing. Next, we built four different machine learning models, including multilayer perceptron (MLP), long short-term memory (LSTM), self-attention LSTM, and self-attention GRU in comparison to the traditional binomial model. The empirical result shows that self-attention GRU with historical data outperforms other models due to its ability to capture complex temporal dependencies and leverage the contextual information embedded in the historical data. Finally, in order to unveil the "black box" of artificial intelligence, we employed the SHapley Additive exPlanations (SHAP) method to interpret and analyze the prediction results of the self-attention GRU model with historical data. This provides insights into the significance and contributions of different input features on the pricing of American-style options.

Pricing of Securities,Machine Learning

What problem does this paper attempt to address?

This paper aims to address the pricing problem of American options. In traditional financial theory, the Black-Scholes model mainly focuses on European options, while American options are more complex to price due to their ability to be exercised at any time before expiration. The paper proposes a machine learning approach that combines Gated Recurrent Unit (GRU) and self-attention mechanism to predict the prices of American options. Firstly, the data is divided into 15 subsets based on the strike price and expiration date, and matching corresponding American Treasury bond rates and implied volatility indices. Then, four machine learning models including Multi-Layer Perceptron (MLP), Long Short-Term Memory (LSTM), self-attention LSTM, and self-attention GRU are constructed and compared with the traditional binary tree model. Experimental results show that the self-attention GRU model driven by historical data performs the best, capturing complex temporal dependencies and utilizing contextual information in historical data. Finally, in order to improve the interpretability of the model, the paper adopts the SHapley Additive exPlanations (SHAP) method to analyze and interpret the prediction results, revealing the influence of different input features on the pricing of American options. This helps understand the working principle of the model and enhance its credibility.

American Option Pricing using Self-Attention GRU and Shapley Value Interpretation

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

Pricing American Options using Machine Learning Algorithms

A Novel Option Pricing Approach Using the Black-Scholes Model and Grey Forecasting Method

Option Pricing Based on HMM and GARCH Model

Improving option price forecasts with neural networks and support vector regressions

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

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

Pricing Options In Hong Kong Market Based On Neural Networks

Deep learning of optimal exercise boundaries for American options

A Gaussian Process Based Method with Deep Kernel Learning for Pricing High-dimensional American Options

American Option Valuation Under the Framework of CGMY Model with Regime-Switching Process

Efficient Estimation of Lower and Upper Bounds for Pricing Higher-Dimensional American Arithmetic Average Options by Approximating Their Payoff Functions

American Call Options Pricing With Modular Neural Networks

Pricing European Options with Google AutoML, TensorFlow, and XGBoost

Financial Option Volatility Prediction Based on Machine Learning Algorithm

Forecasting Implied Volatility Smile Surface via Deep Learning and Attention Mechanism

Building Fuzzy Levy-GJR-GARCH American Option Pricing Model.

American options in time-dependent one-factor models: Semi-analytic pricing, numerical methods and ML support

Prediction of Barrier Option Price Based on Antithetic Monte Carlo and Machine Learning Methods

A Novel Numerical Approach of Computing American Option.