Dice Question Streamline Icon: https://streamlinehq.com

Robust methods for increased market efficiency and complexity

Develop deep learning approaches for Limit Order Book-based stock price trend prediction that maintain robustness and predictive performance under increased market efficiency and complexity, improving generalization from datasets like FI-2010 to more efficient markets such as NASDAQ stocks (e.g., Tesla and Intel).

Information Square Streamline Icon: https://streamlinehq.com

Background

The paper shows that many existing models struggle to generalize beyond the FI-2010 benchmark to more complex and efficient markets like NASDAQ, where predictive performance declines. Even with the proposed architectures, performance on NASDAQ data remains notably lower than on FI-2010.

To address this gap between academic benchmarks and real-world markets, the authors explicitly state that developing more robust approaches to handle increased market efficiency and complexity is still an open question.

References

The investigation of scaling laws for financial deep learning models remains an open question, as does the development of more robust approaches to handling increased market efficiency and complexity.

— TLOB: A Novel Transformer Model with Dual Attention for Price Trend Prediction with Limit Order Book Data (2502.15757 - Berti et al., 12 Feb 2025) in Conclusion (Future works)