Paper: arXiv 2509.07987

Authors: Yiran Wan, Xinyu Ying, Shengzhen Xu

Abstract

Straddle Option is a financial trading tool that explores volatility premiums in high-volatility markets without predicting price direction. Although deep reinforcement learning has emerged as a powerful approach to trading automation in financial markets, existing work mostly focused on predicting price trends and making trading decisions by combining multi-dimensional datasets like blogs and videos, which led to high computational costs and unstable performance in high-volatility markets. To tackle this challenge, we develop automated straddle option trading based on reinforcement learning and attention mechanisms to handle unpredictability in high-volatility markets. Firstly, we leverage the attention mechanisms in Transformer-DDQN through both self-attention with time series data and channel attention with multi-cycle information. Secondly, a novel reward function considering excess earnings is designed to focus on long-term profits and neglect short-term losses over a stop line. Thirdly, we identify the resistance levels to provide reference information when great uncertainty in price movements occurs with intensified battle between the buyers and sellers. Through extensive experiments on the Chinese stock, Brent crude oil, and Bitcoin markets, our attention-based Transformer-DDQN model exhibits the lowest maximum drawdown across all markets, and outperforms other models by 92.5% in terms of the average return excluding the crude oil market due to relatively low fluctuation.

Complexity vs Empirical Score

  • Math Complexity: 5.5/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents an innovative application of Transformer-DDQN combined with self-attention and channel attention to trade straddle options without predicting price direction. It grounds its execution in standard Black-Scholes pricing and resistance level heuristics, demonstrating strong empirical performance across multiple high-volatility asset classes.

Research Flowchart

  flowchart TD
    A[Research Goal: Automated Straddle Option Trading in High-Volatility Markets] --> B[Data Inputs: Time Series & Multi-Cycle Market Data<br><i>Chinese Stocks, Crude Oil, Bitcoin</i>]
    B --> C[Transformer-DDQN Architecture<br><i>Self-Attention + Channel Attention</i>]
    C --> D[Model Enhancements: Resistance Level Identification & Excess Earnings Reward Function]
    D --> E[Outcomes: Lowest Max Drawdown across all markets & 92.5% higher average return in high-volatility markets]

Cited References (85)

  1. Blackboard Systems (1986)
  2. Communication, Simulation, and Intelligent Agents: Implications of Personal Intelligent Machines for Medical Education (1983)
  3. Classification Problem Solving (1984)
  4. New Ways to Make Microcircuits Smaller (1980) — link
  5. New Ways to Make Microcircuits Smaller—Duplicate Entry (1980)
  6. Strategic explanations for a diagnostic consultation system (1984) — link
  7. Strategic Explanations in Consultation—Duplicate (1983)
  8. Poligon: A System for Parallel Problem Solving (1986)
and 77 more
  1. Transfer of Rule-Based Expertise through a Tutorial Dialogue (1979)
  2. The Engineering of Qualitative Models (2021)
  3. Attention Is All You Need (2017) — link
  4. Pluto: The ‘Other’ Red Planet (2015)
  5. Stock Price Forecast Based on LSTM and DDQN (2022) — link
  6. A deep Q-learning based algorithmic trading system for commodity futures markets (2024) — link
  7. Adaptive quantitative trading: An imitative deep reinforcement learning approach (2020) — link
  8. Deep direct reinforcement learning for financial signal representation and trading (2016) — link
  9. Deep reinforcement learning for option replication and hedging (2020) — link
  10. A deep learning based stock trading model with 2-D CNN trend detection (2017) — link
  11. Stock market’s price movement prediction with LSTM neural networks (2017) — link
  12. A LSTM-based method for stock returns prediction: A case study of China stock market (2015) — link
  13. A Na "\i ve SVM-KNN based stock market trend reversal analysis for Indian benchmark indices (2015) — link
  14. Stock price prediction using support vector regression on daily and up to the minute prices (2018) — link
  15. Stock Price Prediction based on LSTM and XGBoost Combination Model (2023) — link
  16. Q-learning (1992)
  17. Playing atari with deep reinforcement learning (2013) — link
  18. Deep reinforcement learning with double q-learning (2016) — link
  19. Continuous control with deep reinforcement learning (2015) — link
  20. Attention is all you need (2017)
  21. Time-series forecasting with deep learning: a survey (2021) — link
  22. Squeeze and excitation networks (2018)
  23. Bilinear attention networks (2018)
  24. Computational modelling of visual attention (2001)
  25. A neurobiological model of visual attention and invariant pattern recognition based on dynamic routing of information (1993)
  26. Spatial transformer networks (2015)
  27. Joint line segmentation and transcription for end-to-end handwritten paragraph recognition (2016)
  28. Attention is all you need (2017)
  29. Hierarchical attention network for document classification (2018)
  30. Self-attention generative adversarial networks (2019)
  31. Stock market forecasting using computational intelligence: A survey (2021)
  32. Mastering the game of go without human knowledge (2017)
  33. Discovering faster matrix multiplication algorithms with reinforcement learning (2022)
  34. Graph attention networks (2017) — link
  35. Bilinear attention networks (2018)
  36. Attentional factorization machines: Learning the weight of feature interactions via attention networks (2017) — link
  37. Stock price prediction based on deep reinforcement learning (2022)
  38. Option,Future,and Other Derivatives (2008)
  39. Volatility Trading (2013)
  40. Options investment management (2022)
  41. Technical Analysis of the Futures Markets (1986)
  42. Technical Analysis of Stock Trends (2007)
  43. New Concepts in Technical Trading Systems (1978)
  44. The Dow Theory (1994)
  45. Rise of the machines: Algorithmic trading in the foreign exchange market (2014)
  46. Algorithmic trading: winning strategies and their rationale (2013)
  47. The Optimal Total Costs for Writing a Straddle (2012)
  48. Stock market prediction and portfolio optimization (2021)
  49. Stock market forecasting using computational intelligence: A survey (2021)
  50. A real time clustering and SVM based price-volatility prediction for optimal trading strategy (2014)
  51. A review of stock market prediction with Artificial neural network (ANN) (2013)
  52. Deep Reinforcement Learning for Trading (2020)
  53. Deep reinforcement learning on a multi-asset environment for trading (2021) — link
  54. An application of deep reinforcement learning to algorithmic trading (2021)
  55. Choosing the right options trading strategy: Riskreturn trade-off and performance in different market conditions (2022)
  56. Case study of event risk management with options strangles and straddles (2022)
  57. Production, hedging, and speculative decisions with options and futures markets (1991)
  58. Hedging volatility risk (2006)
  59. An optimal trading rule under a switchable mean-reversion model (2014)
  60. Momentum and mean reversion across national equity markets (2006)
  61. Improvement and test of stock index futures trading model based on Bollinger bands (2017)
  62. Parameter optimization for trading algorithms of technical agents (2022)
  63. A stock trading algorithm based on trend forecasting and time series classification (2021)
  64. Stock market prediction using machine learning techniques (2021)
  65. Stock Price Prediction based on LSTM and XGBoost Combination Model (2023)
  66. Prediction of stock prices using statistical and machine learning models: a comparative analysis (2022)
  67. Stock market prediction using machine learning classifiers and social media, news (2022)
  68. Stock price prediction using technical indicators: a predictive model using optimal deep learning (2019)
  69. Stock market prediction using data mining techniques (2017)
  70. A synchronous deep reinforcement learning model for automated multi-stock trading (2021)
  71. Optimizing automated trading systems with deep reinforcement learning (2023)
  72. Stock trading bot using deep reinforcement learning (2019)
  73. Deep reinforcement learning approach for trading automation in the stock market (2022)
  74. Stock market prediction using machine learning techniques (2021)
  75. The profitability of moving average trading rules in South Asian stock markets (2001)
  76. Xgboost: A scalable tree boosting system (2016)
  77. Deep reinforcement learning for trading—A critical survey (2021)

0 of 85 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.