Paper: arXiv 2506.05565

Authors: Feliks Bańka, Jarosław A. Chudziak

Abstract

Accurate option pricing is essential for effective trading and risk management in financial markets, yet it remains challenging due to market volatility and the limitations of traditional models like Black-Scholes. In this paper, we investigate the application of the Informer neural network for option pricing, leveraging its ability to capture long-term dependencies and dynamically adjust to market fluctuations. This research contributes to the field of financial forecasting by introducing Informer’s efficient architecture to enhance prediction accuracy and provide a more adaptable and resilient framework compared to existing methods. Our results demonstrate that Informer outperforms traditional approaches in option pricing, advancing the capabilities of data-driven financial forecasting in this domain.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 5.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper involves advanced deep learning architecture (Informer with ProbSparse attention, O(L log L) complexity) and necessary financial data preprocessing/normalization, indicating moderate-to-high mathematical density. Empirical rigor is moderate due to explicit benchmarking against traditional models and profit analysis on historical data, though it lacks code or dataset specifics, suggesting it’s backtest-ready but not fully implemented.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Option Pricing with Informer"] --> B["Data & Input<br>Options Data & Market Volatility"]
  B --> C["Methodology<br>Informer Transformer Architecture"]
  C --> D["Process<br>Long-term Dependency Modeling"]
  D --> E["Outcome<br>Outperforms Black-Scholes & Traditional Models"]
  E --> F["Result<br>Enhanced Prediction Accuracy & Adaptability"]

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 applying the Informer neural network architecture to option pricing, benchmarking it against traditional and machine‑learning models.

Source sentence
We apply the Informer architecture to option pricing, leveraging its long-sequence modeling

Assumes 3

  • Options with a time‑to‑maturity below 30 days are excluded from the dataset. data
    Source sentence
    Options with a time-to-maturity (TTM) below 30 days were excluded, as such short-term contracts are typically highly volatile and speculative (Heston, 1993).
  • Only options with a moneyness ratio between 0.6 and 1.3 are retained. data
    Source sentence
    only options with a moneyness ratio (the ratio of the underlying asset’s price to the strike price) between 0.6 and 1.3 were included, as near-the-money options are more liquid and exhibit more reliable pricing (Bakshi et al., 2000).
  • The train/validation/test split (70/15/15) simulates real‑world conditions where future predictions depend on past training. execution
    Source sentence
    The dataset is split into training, validation, and test sets, with 70% of the data allocated for training, 15% for validation, and the remaining 15% for testing

Data

  • Apple Inc. (AAPL) option contracts · publicly available financial databases options public
    2016-01-04 to 2023-03-31 · Apple options (calls and puts)
    Source sentence
    We use a dataset comprising eight years of historical option contracts for Apple Inc. (AAPL), sourced from publicly available financial databases, covering the period from January 4, 2016, to March 31, 2023.

Costs

Not modeled A trading or execution result is reported without costs.

Finds 2

  • Informer achieves the lowest MAE (2.7145) and RMSE (3.6766) among all tested models. predictive-accuracy
    Source sentence
    Informer 2.7145 3.6766 LSTM 3.9343 5.0373 Black‑Scholes 4.1765 5.3840 Heston 4.1282 5.3565
  • Informer generates the highest cumulative net value in trading evaluations, outperforming Black‑Scholes, Heston, and LSTM. performance
    Source sentence
    The Informer not only achieved the lowest MAE and RMSE across all tested models but also generated the highest cumulative net value in trading evaluations, outperforming all other models...

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 46 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 3 papers in the archive

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 (15)

  1. The Study of Option Pricing Problems based on Transformer Model (2022) — link
  2. Applied Option Pricing using Transformers (2024) — link
  3. The Volatility Surface: A Practitioner’s Guide (2006)
  4. Elements of Financial Risk Management (2009)
  5. Toward Predictive Stock Trading with Hidformer Integrated into Reinforcement Learning Strategy (2024)
  6. Transformer-Style Neural Network in Stock Price Forecasting (2024)
  7. ElliottAgents: A Natural Language-Driven Multi-Agent System for Stock Market Analysis and Prediction (2024)
  8. Predictability of stock returns using neural networks: Elusive in the long term (2023)
and 7 more
  1. Long-Term Rolling Window for Stock Market Predictions (2019)
  2. A deep learning framework for financial time series using stacked autoencoders and long-short term memory (2017) — link
  3. Machine learning approaches in stock market prediction: A systematic literature review (2023) — link
  4. A Closed-Form Solution for Options with Stochastic Volatility with Applications to Bond and Currency Options (1993)
  5. Pricing and hedging long-term options (2000) — link
  6. Neural networks for option pricing and hedging: a literature review (2020) — link
  7. FRACTAL MARKETS HYPOTHESIS AND THE GLOBAL FINANCIAL CRISIS: SCALING, INVESTMENT HORIZONS AND LIQUIDITY (2012) — link

0 of 15 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.