Advanced Neural Network Training Methods for Low False Alarm Stock Trend Prediction

Emad W. Saad; Danil V. Prokhorov; Donald C. Wunsch

doi:10.1109/ICNN.1996.549212

Advanced Neural Network Training Methods for Low False Alarm Stock Trend Prediction

Emad W. Saad
, Danil V. Prokhorov
, Donald C. Wunsch

Research output: Contribution to journal › Article › peer-review

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.

Original language	American English
Journal	Neural Networks, 1996. IEEE International Conference on Neural Networks
Volume	4
DOIs	https://doi.org/10.1109/ICNN.1996.549212
State	Published - Jan 1 1996

Disciplines

Electrical and Computer Engineering

Access to Document

10.1109/ICNN.1996.549212License: Unspecified

https://doi.org/10.1109/ICNN.1996.549212

Cite this

@article{8a389afe60ea455ba45a02239008baad,

title = "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.",

author = "Saad, \{Emad W.\} and Prokhorov, \{Danil V.\} and Wunsch, \{Donald C.\}",

year = "1996",

month = jan,

day = "1",

doi = "10.1109/ICNN.1996.549212",

language = "American English",

volume = "4",

journal = "Neural Networks, 1996. IEEE International Conference on Neural Networks",

}

TY - JOUR

T1 - Advanced Neural Network Training Methods for Low False Alarm Stock Trend Prediction

AU - Saad, Emad W.

AU - Prokhorov, Danil V.

AU - Wunsch, Donald C.

PY - 1996/1/1

Y1 - 1996/1/1

N2 - 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.

AB - 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.

UR - https://scholarsmine.mst.edu/ele_comeng_facwork/284

UR - https://doi.org/10.1109/ICNN.1996.549212

U2 - 10.1109/ICNN.1996.549212

DO - 10.1109/ICNN.1996.549212

M3 - Article

VL - 4

JO - Neural Networks, 1996. IEEE International Conference on Neural Networks

JF - Neural Networks, 1996. IEEE International Conference on Neural Networks

ER -

Advanced Neural Network Training Methods for Low False Alarm Stock Trend Prediction

Abstract

Disciplines

Access to Document

Other files and links

Fingerprint

Cite this