Paper: arXiv 2405.05780

Authors: Daniel de Souza Santos, Tiago Alessandro Espinola Ferreira

Abstract

One of the most discussed problems in the financial world is stock option pricing. The Black-Scholes Equation is a Parabolic Partial Differential Equation which provides an option pricing model. The present work proposes an approach based on Neural Networks to solve the Black-Scholes Equations. Real-world data from the stock options market were used as the initial boundary to solve the Black-Scholes Equation. In particular, times series of call options prices of Brazilian companies Petrobras and Vale were employed. The results indicate that the network can learn to solve the Black-Sholes Equation for a specific real-world stock options time series. The experimental results showed that the Neural network option pricing based on the Black-Sholes Equation solution can reach an option pricing forecasting more accurate than the traditional Black-Sholes analytical solutions. The experimental results making it possible to use this methodology to make short-term call option price forecasts in options markets.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper involves advanced mathematics, including solving the Black-Scholes PDE (a parabolic PDE) and neural network training, justifying a high math score. It uses real-world data from specific stocks, compares results with analytical solutions, and discusses implementation for forecasting, giving it solid empirical rigor.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Develop Neural Network approach to solve Black-Scholes<br>PDE for Option Pricing"] --> B["Data Preparation<br>Time Series: Petrobras & Vale<br>Call Options Prices"]
  B --> C["Model Architecture<br>Feedforward Neural Network<br>Inputs: S, K, T, r, σ<br>Output: Option Price"]
  C --> D["Training Process<br>Learn Black-Scholes Solution<br>from Real Market Data"]
  D --> E["Prediction Output<br>Neural Network Pricing<br>vs<br>Traditional Analytical BS"]
  E --> F["Key Findings<br>NN achieves higher accuracy<br>than Black-Scholes analytical<br>enables short-term forecasting"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a neural network approach to solve the Black‑Scholes equation for option pricing using real‑world option price data.

Source sentence
The present work proposes an approach based on Neural Networks to solve the Black‑Scholes Equations.

Assumes 4

  • Only call options are used for training because put options have low trading volume. data
    Source sentence
    As a case study, given data availability limitations, only the call option series were used for the NN training. The puts did not have significant trading volume.
  • The option price data are free of charge and obtained from the ADVFN website. data
    Source sentence
    Options price data can be obtained free of charge on the ADVFN website.
  • The neural network has two inputs: a modified price (x) and a time interval (τ). model
    Source sentence
    since the equation involves two variables, one represents a modified price (x) and the other represents an interval of time (τ). Thus, the network has two inputs.
  • All option series have the same size of 61 data points. data
    Source sentence
    In this data set all series have the same size of 61 points.

Data

  • Petrobras and Vale call option series · ADVFN options public
    Petrobras (PETR4) and Vale (VALE3) stocks
    Source sentence
    Options were selected concerning the two most traded stocks on the Brazilian Stock Exchange: Petrobras (PETR4) and Vale (VALE3).

Costs

Not applicable The paper reports no trading or execution result.

Finds 4

  • The neural network achieved the lowest MAE of 0.670 and MSE of 0.751 for the PETRD266 option. predictive-accuracy
    Source sentence
    The NN price prediction for PETRD266 option (K = 27.76 BRL) showed the lowest values for MAE (0.670) and MSE (0.751).
  • For the PETRD122 option the network recorded the lowest ARV (0.018) and the highest POCID (35.000). predictive-accuracy
    Source sentence
    the NN solution for call PETRD122 (K = 22.26 BRL) had the lower recorded value of ARV (0.018) and the highest POCID (35.000).
  • The best MAE for Petrobras options (≈2.6) is about seven times lower than the MAE (14.3) reported in a comparable study. performance
    Source sentence
    Our work presented the best scenario when studying options on Petrobras, whose MAE was about seven times lower, around 2.6.
  • For Vale time series the neural network achieved an average MAE of 5.6. predictive-accuracy
    Source sentence
    When looking at the VALE times series, our NN prediction reached an average MAE of 5.6.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 21 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 2 papers in the archive (2 by independent authors)

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (42)

  1. Static hedging and pricing American knock-in put options (2012) — link
  2. A New Intelligent System Methodology for Time Series Forecasting with Artificial Neural Networks (2008) — link
  3. Improving option price forecasts with neural networks and support vector regressions (2009) — link
  4. The Pricing of Options and Corporate Liabilities (1973) — link
  5. Artificial neural networks for solving ordinary and partial differential equations (1998) — link
  6. Hedging of the Barrier Put Option in a Diffusion (B, S) – Market in case of Dividends Payment on a Risk Active (2015) — link
  7. Introduction to the Mathematics of Finance (2004) — link
  8. NeuroDiffEq: A Python package for solving differential equations with neural networks (2020) — link
and 34 more
  1. Introductory Mathematical Analysis for Quantitative Finance (2020) — link
  2. Pricing options with dual volatility input to modular neural networks (2020) — link
  3. An efficient numerical method for pricing American put options under the CEV model (2020) — link
  4. Hermite polynomial based expansion of European option prices (2014) — link
  5. Calibration of the double Heston model and an analytical formula in pricing American put option (2021) — link
  6. Learning sequential option hedging models from market data (2021) — link
  7. The Effect of Option-implied Skewness on Delta- and Vega-Hedged Option Returns (2019) — link
  8. Time-sequencing European options and pricing with deep learning – Analyzing based on interpretable ALE method (2021) — link
  9. Penalty method for indifference pricing of American option in a liquidity switching market (2021) — link
  10. Option Pricing Based on GA-BP neural network (2022) — link
  11. Pricing discounted American capped options (2022) — link
  12. The impact of option hedging on the spot market volatility (2022) — link
  13. A soft robust methodology to devise hedging strategies in renewable energy trading based on electricity options (2022) — link
  14. Derivative Pricing (2018) — link
  15. Pandemic portfolio choice (2022) — link
  16. An accurate and stable numerical method for option hedge parameters (2022) — link
  17. Pricing American options with stochastic volatility and small nonlinear price impact: A PDE approach (2022) — link
  18. Parameter estimation of uncertain differential equation by implementing an optimized artificial neural network (2022) — link
  19. Numerical Techniques for Determining Implied Volatility in Option Pricing (2022) — link
  20. GARCH option pricing with volatility derivatives (2022) — link
  21. Neural network stochastic differential equation models with applications to financial data forecasting (2022) — link
  22. Fractional Liu uncertain differential equation and its application to finance (2022) — link
  23. Parallel computing in finance for estimating risk-neutral densities through option prices (2022) — link
  24. American options pricing under regime-switching jump-diffusion models with meshfree finite point method (2022) — link
  25. Option price implied information and REIT returns (2023) — link
  26. Hedging performance using google Trends–Evidence from the indian forex options market (2023) — link
  27. A physics-informed neural network based on mixed data sampling for solving modified diffusion equations (2023) — link
  28. Wavelets based physics informed neural networks to solve non-linear differential equations (2023) — link
  29. Greedy training algorithms for neural networks and applications to PDEs (2023) — link
  30. Neural ordinary differential equation for sequential optimal design of fatigue test under accelerated life test analysis (2023) — link
  31. Mathematical modeling of the unemployment problem in a context of financial crisis (2023) — link
  32. A novel physics-informed framework for real-time adaptive modal parameters estimation of offshore structures (2023) — link
  33. Physics-informed variational inference for uncertainty quantification of stochastic differential equations (2023) — link
  34. Reliable Extrapolation of Deep Neural Operators Informed by Physics or Sparse Observations (2023) — link

0 of 42 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.