Stock Price Prediction Using Temporal Graph Model with Value Chain Data

Chang LiuSandra Paterlini

Mar 2023

摘要

Stock price prediction is a crucial element in financial trading as it allowstraders to make informed decisions about buying, selling, and holding stocks.Accurate predictions of future stock prices can help traders optimize theirtrading strategies and maximize their profits. In this paper, we introduce aneural network-based stock return prediction method, the Long Short-Term MemoryGraph Convolutional Neural Network (LSTM-GCN) model, which combines the GraphConvolutional Network (GCN) and Long Short-Term Memory (LSTM) Cells.Specifically, the GCN is used to capture complex topological structures andspatial dependence from value chain data, while the LSTM captures temporaldependence and dynamic changes in stock returns data. We evaluated the LSTM-GCNmodel on two datasets consisting of constituents of Eurostoxx 600 and S&P 500.Our experiments demonstrate that the LSTM-GCN model can capture additionalinformation from value chain data that are not fully reflected in price data,and the predictions outperform baseline models on both datasets.

展开全部

图表提取

论文十问

参考文献

被引用

提问与回答

暂无人提供速读十问回答

被引用

笔记

问答

领域