PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 9, 20260 citationsOpen Access

Hierarchical Attention Fused CNN-LSTM Using Structured 2D Indicator Matrices for Stock Trading Action Detection

View Full Paper
HFHao FengXLXian LiDZDongjie Zhao

Key Points

  • The aim is to improve the detection of trading actions like buy, sell, and hold in stock markets by addressing existing limitations.
  • Developed a deep learning framework named HA-CL that combines CNN and LSTM architectures.
  • Encoded technical indicators into structured 2D matrices for effective feature representation.
  • Implemented a hierarchical attention mechanism to enhance feature importance at multiple levels.
  • Designed a customized extrema labeling strategy along with oversampling techniques to address class imbalance.
  • Utilized importance-aware focal loss and heuristic action recalibration to improve model robustness.
  • Achieved an average accuracy of 68.89% in detecting trading actions.
  • Secured an annualized return of 111.01%, indicating strong trading performance.
  • Demonstrated superior performance against baseline models and validated using risk-adjusted return metrics.

Abstract

Accurate detection of trading actions (buy, sell, and hold) is critical for portfolio optimization and risk management in volatile stock markets. However, existing approaches often suffer from deficiencies in feature representation, spatiotemporal modeling, and class balancing, which limit their effectiveness. To address these issues, we propose HA-CL, a deep learning framework that integrates a hierarchical attention mechanism with CNN-LSTM. Specifically, technical indicators are encoded into a structured 2D matrix to preserve the inherent characteristics of stocks. Features extracted by ResNet are processed by a channel-wise LSTM equipped with an attention core to adaptively fuse spatial, temporal, and channel-level importance. To mitigate class imbalance, we design a customized extrema labeling strategy augmented with extrema oversampling, an importance-aware focal loss, and a heuristic action recalibration. Experiments on 63 Chinese A-share stocks show that HA-CL achieves an average accuracy of 68.89% with an annualized return of 111.01%, substantially outperforming all baselines. Risk-adjusted return metrics such as the Sharpe Ratio and the Maximum Drawdown further validate its robustness across market conditions. Together, they highlight the potential of HA-CL to translate complex market patterns into profitable trading actions.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Feng et al. (2026) studied this question.

synapsesocial.com/papers/69897a35f0ec2af6756e890dhttps://doi.org/10.3390/app16041672
Ask AI
Helpful
Bookmark
Share
View Full Paper