TY - JOUR
T1 - Empirical validation of ELM trained neural networks for financial modelling
AU - Novykov, Volodymyr
AU - Bilson, Christopeher M
AU - Gepp, Adrian
AU - Harris, Geoffrey
AU - Vanstone, Bruce J
N1 - Funding Information:
The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This work was supported by the Australian Government Department of Education and Training through the Australian Government Research Training Program (RTP) Scholarship to Volodymyr Novykov.
Publisher Copyright:
© 2022, The Author(s).
PY - 2023/1
Y1 - 2023/1
N2 - The purpose of this work is to compare predictive performance of neural networks trained using the relatively novel technique of training single hidden layer feedforward neural networks (SFNN), called Extreme Learning Machine (ELM), with commonly used backpropagation-trained recurrent neural networks (RNN) as applied to the task of financial market prediction. Evaluated on a set of large capitalisation stocks on the Australian market, specifically the components of the ASX20, ELM-trained SFNNs showed superior performance over RNNs for individual stock price prediction. While this conclusion of efficacy holds generally, long short-term memory (LSTM) RNNs were found to outperform for a small subset of stocks. Subsequent analysis identified several areas of performance deviations which we highlight as potentially fruitful areas for further research and performance improvement.
AB - The purpose of this work is to compare predictive performance of neural networks trained using the relatively novel technique of training single hidden layer feedforward neural networks (SFNN), called Extreme Learning Machine (ELM), with commonly used backpropagation-trained recurrent neural networks (RNN) as applied to the task of financial market prediction. Evaluated on a set of large capitalisation stocks on the Australian market, specifically the components of the ASX20, ELM-trained SFNNs showed superior performance over RNNs for individual stock price prediction. While this conclusion of efficacy holds generally, long short-term memory (LSTM) RNNs were found to outperform for a small subset of stocks. Subsequent analysis identified several areas of performance deviations which we highlight as potentially fruitful areas for further research and performance improvement.
UR - http://www.scopus.com/inward/record.url?scp=85139196155&partnerID=8YFLogxK
U2 - 10.1007/s00521-022-07792-3
DO - 10.1007/s00521-022-07792-3
M3 - Article
SN - 0941-0643
VL - 35
SP - 1581
EP - 1605
JO - Neural Computing and Applications
JF - Neural Computing and Applications
IS - 2
ER -