Integrated GCN–BiGRU–TPE Agricultural Product Futures Prices Prediction Based on Multi-graph Construction

Dabin Zhang, Xiaoming Li (), Liwen Ling, Huanling Hu and Ruibin Lin
Additional contact information
Dabin Zhang: South China Agricultural University
Xiaoming Li: South China Agricultural University
Liwen Ling: South China Agricultural University
Huanling Hu: South China Agricultural University
Ruibin Lin: South China Agricultural University

Computational Economics, 2025, vol. 66, issue 5, No 11, 3927-3955

Abstract: Abstract Accurate prediction of agricultural product futures prices is crucial for the sustainable and healthy development of the agricultural industry. Existing neural network models for predicting agricultural product futures prices have not fully considered the nonlinear correlation and structural relationship of input data, which may limit the model's capability to capture and predict price dynamics. Therefore, this paper proposes a combined prediction model that integrates graph convolutional network (GCN) with bidirectional gated recurrent unit (BiGRU) and integrates them through tree-structured parzen estimator (TPE). The GCN–BiGRU–TPE model offers more comprehensive and in-depth data analysis capabilities. From the spatial dimension, the GCN captures the complex topological structure of the data; from the temporal dimension, the BiGRU processes the time dependency of price sequences. The TPE optimizes the weights of the spatiotemporal features outputted, which are then passed through a fully connected layer for final prediction. Empirical research on corn futures price data shows that the proposed GCN–BiGRU–TPE model outperforms traditional prediction models. In various error metrics used, the root mean square error (RMSE) was 20.214, the mean absolute error (MAE) was 14.870, the mean absolute percentage error (MAPE) was 0.547, and the r-squared ( $${\text{R}}^{2}$$ R 2 ) was 0.969. These results highlight the effectiveness of applying graph-structured data and graph neural networks in predicting agricultural product prices.

Keywords: Agricultural futures forecasting; Deep learning; Graph neural network; Distance correlation coefficient; Multivariate time series (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10614-024-10832-w Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:kap:compec:v:66:y:2025:i:5:d:10.1007_s10614-024-10832-w

Ordering information: This journal article can be ordered from
http://www.springer. ... ry/journal/10614/PS2

DOI: 10.1007/s10614-024-10832-w

Access Statistics for this article

Computational Economics is currently edited by Hans Amman

More articles in Computational Economics from Springer, Society for Computational Economics Contact information at EDIRC.
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().