Paper: arXiv 2407.03760
Authors: Yuhui Jin
Abstract
The application of deep learning techniques for predicting stock market prices is a prominent and widely researched topic in the field of data science. To effectively predict market trends, it is essential to utilize a diversified dataset. In this paper, we give a graph neural network based convolutional neural network (CNN) model, that can be applied on diverse source of data, in the attempt to extract features to predict the trends of indices of \text{“S”}&\text{“P”} 500, NASDAQ, DJI, NYSE, and RUSSEL. The experiments show that the associated models improve the performance of prediction in all indices over the baseline algorithms by about $4% \text{" to “} 15%$, in terms of F-measure. A trading simulation is generated from predictions and gained a Sharpe ratio of over 3.
Complexity vs Empirical Score
- Math Complexity: 7.0/10
- Empirical Rigor: 6.0/10 (capped from 8.0: transaction costs not modelled; see penalty flags)
- Quadrant: Holy Grail — high math complexity, high empirical rigor (effective empirical rigor capped from 8.0)
Why this score: The paper introduces advanced graph neural network architectures (GCN, GAT) with convolutional layers and mathematical formulations for message passing, indicating high math complexity. It also demonstrates empirical rigor by testing on multiple indices, reporting specific performance metrics (F-measure improvements, Sharpe ratio >3), and detailing a trading simulation.
Research Flowchart
flowchart TD A["Research Goal: Predict stock market indices<br>using diverse data sources with a<br>Graph Neural Network & CNN model"] --> B["Data Collection & Graph Construction"] B --> C["Data: S&P 500, NASDAQ, DJI, NYSE, RUSSEL<br>Input: Historical Price & Volume Data"] C --> D["Model Architecture: GraphCNNpred"] D --> E["Graph Convolution & Feature Extraction"] E --> F["Time Series Prediction<br>(Trend Prediction)"] F --> G["Trading Simulation<br>(Sharpe Ratio Analysis)"] G --> H["Key Outcomes: 4-15% Improvement in F-Measure<br>Sharpe Ratio > 3 vs Baseline Algorithms"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes a graph neural network based convolutional neural network (Graph-CNNpred) that combines GNNs with CNNs for stock index trend prediction.
Source sentence
we give a graph neural network based convolutional neural network (CNN) model, that can be applied on diverse source of data, in the attempt to extract features to predict the trends of indices of
Assumes 2
- No edge information is allowed in the GNN layers. model
Source sentence
An extra assumption is imposed here for the rest of this section: no edge information is allowed.
- Transaction costs are ignored in the trading simulation. execution
Source sentence
Transaction costs are excluded in this analysis.
Data
S&P 500, DJI, NASDAQ, NYSE, RUSSELL indices · equities · daily
- Dataset from [9] macro unknown
Source sentence
The dataset covers the period from January 2010 to November 2017, and we split it into training (65%), validation (15%), and test (20%) sets.
Costs
Not modeled
Source sentence
Transaction costs are excluded in this analysis.
Finds 2
- Graph-CNNpred improves prediction performance by about 4% to 15% in F-measure across all five indices compared to baselines. predictive-accuracy
Source sentence
Across all five indices, Graph-CNNpred enhanced prediction performance by approximately 4% to 15%in terms of F-measure.
- The trading simulation based on Graph-CNNpred predictions achieved a Sharpe ratio of over 3. performance
Source sentence
A trading simulation is generated from predictions and gained a Sharpe ratio of over 3.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 19 related papers). Coverage is partial, so this is neither replication nor refutation.
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.
Cited References (18)
- Semi-Supervised Classification with Graph Convolutional Networks (2017) — link
- HATS: A Hierarchical Graph Attention Network for Stock Movement Prediction (2019) — link
- Semi-Supervised Classification with Graph Convolutional Networks (2017) — link
- Encyclopedia of candlestick charts (2012)
- Optimal versus naive diversification: How inefficient is the 1/N portfolio strategy? (2009)
- Intraday prediction of Borsa Istanbul using convolutional neural networks and feature correlations (2017)
- Understanding Pooling in Graph Neural Networks (2024) — link
- Stock Market Prediction via Deep Learning Techniques: A Survey — → 2212.12717
and 10 more
- Knowledge-driven stock trend prediction and explanation via temporal convolutional network (2019)
- U-cnnpred: A universal cnn-based predictor for stock markets (2019) — link
- Using structured events to predict stock price movement: An empirical investigation (2014)
- Deep learning for event-driven stock prediction (2015)
- Mg-conv: A spatiotemporal multi-graph convolutional neural network for stock market index trend prediction (2022)
- Modeling the stock relation with graph network for overnight stock movement prediction (2021)
- Exploring graph neural networks for stock market predictions with rolling window analysis (2019) — link
- Temporal relational ranking for stock prediction (2019)
- Graph Attention Networks (2018) — link
- CNNpred: CNN-based stock market prediction using a diverse set of variables (2019)
1 of 18 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.