Abstract
The foreign exchange market (forex) is globally the most liquid financial market which creates opportunity for traders and investors to make profit through the fluctuations in currency prices. However, the prediction of forex prices is a complex issue, having been classified among the computational hard problems with the assertion that most of the forex markets cannot be predicted. The breakthrough of deep learning in many areas applications in the last decade has suggested that deep learning network models could offer some solutions to the problem of effective prediction of forex prices. The paper is the study of the application of recurrent neural network (RNN), multilayer perceptron (MLP) and long short-term memory (LSTM) for the prediction of forex prices, focusing on the effect of properties that have not been considered in most of the previous studies. The feasibility, practicality and the effect of the application of these network models for future prediction after the training phase are explored. The results of this research are reported.
Presenters
David Ademola OyemadeSenior Lecturer, Computer Science, Federal University of Petroleum Resources, Effurun, Delta, Nigeria
Details
Presentation Type
Paper Presentation in a Themed Session
Theme
KEYWORDS
Deep Learning, Recurrent Neural Network, Multilayer Perceptron, Long Short-Term Memory, Forex Prices Prediction
Digital Media
This presenter hasn’t added media.
Request media and follow this presentation.