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

• URL: http://arxiv.org/abs/2607.19385v1
• Date: Wed, 01 Jul 2026 13:05:33 GMT
• Title: STN-TGAT: Top-K Portfolio Construction via Prior-Guided Graph Attention with Learnable Soft-Threshold Sparsification
• Abstract summary: This paper tackles the problem of stock ranking and portfolio construction under realistic investment settings by jointly modeling temporal dynamics and cross-sectional dependencies.<n>We propose the Soft-Threshold NMI-prior Transformer Graph Attention Network (STN-TGAT), which integrates a temporal Transformer with a Graph Attention Network to capture long-horizon sequential patterns and dynamic inter-stock relationships.<n> Empirical results on real-world data demonstrate that STN-TGAT consistently outperforms benchmark models from predictive accuracy and investment profitability measured by portfolio returns.
• Score: 14.921992059477367
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper tackles the problem of stock ranking and portfolio construction under realistic investment settings by jointly modeling temporal dynamics and cross-sectional dependencies. We propose the Soft-Threshold NMI-prior Transformer Graph Attention Network (STN-TGAT), which integrates a temporal Transformer with a Graph Attention Network to capture long-horizon sequential patterns and dynamic inter-stock relationships. An NMI-based prior graph combined with a soft-threshold sparsification mechanism enhances structural robustness by mitigating noisy correlations while preserving informative connections. The portfolio formation process incorporates practical considerations, including Top-5 selection within the Top-50 $S\&P$ 500 constituents, explicit weight allocation, and transaction cost adjustment, thereby aligning the evaluation with real-world trading conditions. Empirical results on real-world data demonstrate that STN-TGAT consistently outperforms benchmark models from predictive accuracy and investment profitability measured by portfolio returns. These findings suggest that combining decision-aligned training with adaptive relational modeling provides a coherent and practically effective framework for data-driven portfolio construction.

Related papers

• Do Better Volatility Forecasts Lead to Better Portfolios? Evidence from Graph Neural Networks [0.0]
  This paper tests whether graph neural networks improve realized volatility forecasts and whether those forecasts improve portfolio performance.<n>Using weekly realized volatility for 465 S&P 500 equities from 2015-2025, Heterogeneous Autoregressive and Long Short-Term Memory baselines are compared against GraphSAGE models built on rolling correlation, sector, and Granger-causal graphs, with and without macro regime features.<n>The empirical finding is that the model with the lowest forecast MSE, the model with the highest cross-sectional ranking accuracy, and the model with the highest portfolio Sharpe ratio are three different models.
  arXiv  Detail & Related papers  (2026-05-19T02:52:00Z)
• GAPNet: Plug-in Jointly Learning Task-Specific Graph for Dynamic Stock Relation [3.7582774908258294]
  We propose GAPNet, a Graph Adaptation Plug-in Network that jointly learns task-specific topology and representations in an end-to-end manner.<n>Across two real-world stock datasets, GAPNet has been shown to consistently enhance the profitability and stability in comparision to the state-of-the-art models.
  arXiv  Detail & Related papers  (2026-01-31T20:25:25Z)
• Parallel and Multi-Stage Knowledge Graph Retrieval for Behaviorally Aligned Financial Asset Recommendations [46.90931293070464]
  This paper introduces RAG-FLARKO, a retrieval-augmented extension to FLARKO.<n>It overcomes scalability and relevance challenges using multi-stage and parallel KG retrieval processes.<n> Empirical evaluation on a real-world financial transaction dataset demonstrates that RAG-FLARKO significantly enhances recommendation quality.
  arXiv  Detail & Related papers  (2025-10-08T20:42:53Z)
• Estimating Time Series Foundation Model Transferability via In-Context Learning [74.65355820906355]
  Time series foundation models (TSFMs) offer strong zero-shot forecasting via large-scale pre-training.<n>Fine-tuning remains critical for boosting performance in domains with limited public data.<n>We introduce TimeTic, a transferability estimation framework that recasts model selection as an in-context-learning problem.
  arXiv  Detail & Related papers  (2025-09-28T07:07:13Z)
• Dependency Network-Based Portfolio Design with Forecasting and VaR Constraints [8.107171581224312]
  This study proposes a novel portfolio optimization framework that integrates statistical social network analysis with time series forecasting and risk management.<n>Using daily stock data from the S&P 500 ( 2020-2024), we construct dependency networks via Vector Autoregression ( VAR) and Forecast Error Variance Decomposition (FEVD)<n>FEVD breaks down the VAR's forecast error variance to quantify how much each stock's shocks contribute to another's uncertainty information we invert to form influence-based edge weights in our network.<n>A dynamic portfolio is constructed using the top-ranked stocks, with capital allocated based on Value at Risk (
  arXiv  Detail & Related papers  (2025-07-26T18:53:39Z)
• Towards Improving Long-Tail Entity Predictions in Temporal Knowledge Graphs through Global Similarity and Weighted Sampling [53.11315884128402]
  Temporal Knowledge Graph (TKG) completion models traditionally assume access to the entire graph during training.<n>We present an incremental training framework specifically designed for TKGs, aiming to address entities that are either not observed during training or have sparse connections.<n>Our approach combines a model-agnostic enhancement layer with a weighted sampling strategy, that can be augmented to and improve any existing TKG completion method.
  arXiv  Detail & Related papers  (2025-07-25T06:02:48Z)
• A Pre-training Framework for Relational Data with Information-theoretic Principles [57.93973948947743]
  We introduce Task Vector Estimation (TVE), a novel pre-training framework that constructs supervisory signals via set-based aggregation over relational graphs.<n>TVE consistently outperforms traditional pre-training baselines.<n>Our findings advocate for pre-training objectives that encode task heterogeneity and temporal structure as design principles for predictive modeling on relational databases.
  arXiv  Detail & Related papers  (2025-07-14T00:17:21Z)
• A Study of Dynamic Stock Relationship Modeling and S&P500 Price Forecasting Based on Differential Graph Transformer [4.6028394466086535]
  We propose a Differential Graph Transformer framework for dynamic relationship modeling and price prediction.<n>Our DGT integrates sequential graph structure changes into multi-head self-attention.<n>Clausal temporal attention captures global/local dependencies in price sequences.
  arXiv  Detail & Related papers  (2025-06-23T14:53:31Z)
• RelGNN: Composite Message Passing for Relational Deep Learning [56.48834369525997]
  We introduce RelGNN, a novel GNN framework specifically designed to leverage the unique structural characteristics of the graphs built from relational databases.<n>RelGNN is evaluated on 30 diverse real-world tasks from Relbench (Fey et al., 2024), and achieves state-of-the-art performance on the vast majority tasks, with improvements of up to 25%.
  arXiv  Detail & Related papers  (2025-02-10T18:58:40Z)
• Large-scale Time-Varying Portfolio Optimisation using Graph Attention Networks [4.2056926734482065]
  This study utilise 30 years of data on mid-cap firms, creating graphs of firms using distance correlation and the Triangulated Maximally Filtered Graph approach.<n>We show that the portfolio produced by the GAT-based model outperforms all benchmarks and is consistently superior to other strategies over a long period.
  arXiv  Detail & Related papers  (2024-07-22T10:50:47Z)
• EasyDGL: Encode, Train and Interpret for Continuous-time Dynamic Graph Learning [92.71579608528907]
  This paper aims to design an easy-to-use pipeline (termed as EasyDGL) composed of three key modules with both strong ability fitting and interpretability. EasyDGL can effectively quantify the predictive power of frequency content that a model learn from the evolving graph data.
  arXiv  Detail & Related papers  (2023-03-22T06:35:08Z)
• Tensor Graph Convolutional Networks for Multi-relational and Robust Learning [74.05478502080658]
  This paper introduces a tensor-graph convolutional network (TGCN) for scalable semi-supervised learning (SSL) from data associated with a collection of graphs, that are represented by a tensor. The proposed architecture achieves markedly improved performance relative to standard GCNs, copes with state-of-the-art adversarial attacks, and leads to remarkable SSL performance over protein-to-protein interaction networks.
  arXiv  Detail & Related papers  (2020-03-15T02:33:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.