Back to feed
arXiv cs.LG
arXiv cs.LG
7/23/2026
STN-TGAT: Top-K Portfolio Construction via Prior-Guided Graph Attention with Learnable Soft-Threshold Sparsification

STN-TGAT: Top-K Portfolio Construction via Prior-Guided Graph Attention with Learnable Soft-Threshold Sparsification

Short summary

STN-TGAT combines a temporal Transformer with a Graph Attention Network and NMI-based prior graph for stock ranking and portfolio construction. It uses soft-threshold sparsification to filter noisy correlations while preserving informative connections. Evaluated on S&P 500 constituents with Top-5 selection and transaction cost adjustment, it consistently outperforms benchmark models in both predictive accuracy and portfolio returns.

  • •STN-TGAT integrates temporal Transformer with Graph Attention Network for stock ranking and portfolio construction
  • •NMI-based prior graph with soft-threshold sparsification mitigates noisy correlations
  • •Outperforms benchmarks on S&P 500 data with realistic trading conditions including transaction costs

Generated with AI, which can make mistakes.

#research-breakthrough#ai-tools
Read full article at arXiv cs.LG

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

STKAN: Kolmogorov-Arnold Networks for Spatio-Temporal Forecasting
arXiv cs.LG
STAGformer: A Spatio-temporal Agent Graph Transformer for Micro Mobility Demand Forecasting
arXiv cs.LG
Reinforcement Learning-Guided NSGA-II Enhanced with Gray Relational Coefficient for Multi-Objective Optimization: Application to NASDAQ Portfolio Optimization
arXiv cs.LG
Deep Reinforcement Learning for Reliability Based Bi-Objective Portfolio Optimization
arXiv cs.LG