Journal of Engineering, Project, and Production Management, 2026, 16(5), 2025-305
Developing A Stock Prediction Model Using Multi-Source Data and Temporal Encoding
Associate Professor, School of Economics, Liaoning University of International Business and Economics, Dalian, 116000, China, E-mail: renxuxu@126.com
Project Management
Received December 5, 2025; revised January 21, 2026; accepted March 30, 2026
Available online June 17, 2026
Abstract: As a core component of the economic system, the stock market plays a crucial role in asset growth for investors, corporate financing decisions, and the stability of financial markets. Traditional prediction models face issues such as reliance on static weights for multi-source heterogeneous data integration and incomplete capture of temporal features, which limit their practical applicability. Therefore, this study proposes a stock prediction model based on multi-source data and temporal encoding. The model first constructs a multi-source data extraction algorithm using Empirical Mode Decomposition (EMD) and a Text Convolutional Neural Network (TextCNN). On this basis, the model combines Bidirectional Gated Recurrent Units and temporal encoding to predict stock prices. Experimental results show that the proposed model outperforms the comparison model in predicting A-share and H-share stocks. The coefficients of determination for Geely and Dongfeng Motor stocks are 0.925 and 0.918, respectively. All indicators outperform comparison models. The proposed model balances prediction accuracy and stability and provides a quantitative basis for investor strategy formulation, financial institutions risk management, and regulatory decision-making.
Keywords: Multi-source data, temporal encoding, stock prediction, bigru, time2vec.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Ren, X. (2026). Developing A Stock Prediction Model Using Multi-Source Data and Temporal Encoding. Journal of Engineering, Project, and Production Management, 16(5), 2025-305.
