International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | Cross-Market Prediction Using Dynamic Graph Neural Networks |
|---|---|
| ABSTRACT | This research presents a Dynamic Graph Neural Network (DGNN) framework engineered for forecasting volatility and market trends across multiple financial markets. In contrast to conventional models that examine markets separately and ignore interconnections between them, our methodology dynamically identifies the changing relationships among worldwide financial indices using Graph Attention Networks (GAT), integrating both time-based and spatial elements. Our proposed architecture outperforms standard models including LSTM and GCN regarding precision and consistency in volatility forecasting under diverse market scenarios. The results highlight the capability of DGNNs for flexible and explainable financial prediction. |
| AUTHOR | SANDHYA G, M V JAGANNATHA REDDY, TEJAS H R, DARSHAN B C, RAKESH G, VINAY KUMAR P Dept. of Artificial Intelligence and Machine Learning, M S Engineering College, Bengaluru, India |
| VOLUME | 182 |
| DOI | DOI: 10.15680/IJIRCCE. 2026.1403102 |
| pdf/102_Cross-Market Prediction Using Dynamic Graph Neural Networks.pdf | |
| KEYWORDS | |
| References | [1] L. A. Teixeira and A. L. I. De Oliveira present a technique for automated stock trading that integrates technical analysis with nearest neighbor classification in their work published in Expert Syst. Appl., vol. 37, no. 10, pp. 6885–6890, Oct. 2010 [2] J. L. Ticknor discusses a Bayesian regularized artificial neural network designed for predicting stock market trends in Expert Syst. Appl., vol. 40, no. 14, pp. 5501–5506, Oct. 2013 [3] T. N. Kipf and M. Welling, “Semi-Supervised Classification with Graph Convolutional Networks,” presented at the International Conference on Learning Representations (ICLR), 2017. [4] T. Bollerslev, “Generalized Autoregressive Conditional Heteroskedasticity,” published in the Journal of Econometrics, , vol. 31,pp. 307–327, 1986 [5] R. C. Cavalcante, R. C. Brasileiro, V. L. F. Souza, J. P. Nobrega, and A. L. I. Oliveira provide a survey and future directions on computational intelligence in financial markets in Expert Syst. Appl., vol. 55, pp. 194–211, Aug. 2016 [6] B. Vanstone and G. Finnie propose an empirical approach for creating stock market trading systems using artificial neural networks, as found in Expert Syst. Appl., vol. 36, no. 3, pp. 6668–6680, Apr. 2009 [7] J. J. Murphy authored Technical Analysis of the Financial Markets: A Comprehensive Guide to Trading Methods and Applications, released by the New York Institute of Finance in 1999 [8] D. M. Q. Nelson, A. C. M. Pereira, and R. A. de Oliveira examine stock market price movement prediction through LSTM neural networks in the proceedings of the Int. Joint Conf. Neural Netw. (IJCNN), Anchorage, AK, USA, May 2017, pp. 1419–1426 [9] W. Bao, J. Yue, and Y. Rao introduce a deep learning framework for financial time series analysis employing stacked autoencoders and long-short term memory in PLoS ONE, vol. 12, no. 7, Jul. 2017, Art. no. e0180944. [10] J. Wang, T. Sun, B. Liu, Y. Cao, and D. Wang explore financial market predictions using deep learning in the proceedings of the 17th IEEE Int. Conf. Mach. Learn. Appl. (ICMLA), Dec. 2018, pp. 97–104. [11] G. S. Atsalakis and K. P. Valavanis provide an overview of stock market forecasting techniques, emphasizing soft computing approaches, in Expert Syst. Appl., vol. 36, no. 3, pp. 5932–5941, Apr. 2009. |
