Advanced Neural Network Training Methods for Low False Alarm Stock Trend Prediction
Abstract
Two possible neural network architectures for stock market forecasting are the time-delay neural network and the recurrent neural network. In this paper we explore two effective techniques for training of the above networks, i.e. conjugate gradient algorithm and multi-stream extended Kalman filter. We are particularly interested in limiting false alarms, which corresponds to actual investment losses. Encouraging results have been obtained when using the above techniques.
Recommended Citation
E. W. Saad et al., "Advanced Neural Network Training Methods for Low False Alarm Stock Trend Prediction," Neural Networks, 1996. IEEE International Conference on Neural Networks, vol. 4, pp. 2021 - 2026, Institute of Electrical and Electronics Engineers (IEEE), Jan 1996.
The definitive version is available at https://doi.org/10.1109/ICNN.1996.549212
Meeting Name
1996 IEEE International Conference on Neural Networks, ICNN (1996: Jun. 3-6, Washington, DC)
Department(s)
Electrical and Computer Engineering
International Standard Book Number (ISBN)
0000780332105
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 1996 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.
Publication Date
01 Jan 1996