Assess deep learning models for Egyptian exchange prediction using nonlinear artificial neural networks
- PDF / 1,189,695 Bytes
- 23 Pages / 595.276 x 790.866 pts Page_size
- 71 Downloads / 197 Views
(0123456789().,-volV)(0123456789(). ,- volV)
ORIGINAL ARTICLE
Assess deep learning models for Egyptian exchange prediction using nonlinear artificial neural networks Essam H. Houssein1
•
Mahmoud Dirar1 • Kashif Hussain2 • Waleed M. Mohamed1
Received: 6 April 2020 / Accepted: 18 September 2020 Ó Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract Financial analysis of the stock market using the historical data is the exigent demand in business and academia. This work explores the efficiency of three deep learning (Dl) techniques, namely Bayesian regularization (BE), Levenberg–Marquardt (lM), and scaled conjugate gradient (SCG), for training nonlinear autoregressive artificial neural networks (NARX) for predicting specifically the closing price of the Egyptian Stock Exchange indices (EGX-30, EGX-30-Capped, EGX-50EWI, EGX-70, EGX-100, and NIlE). An empirical comparison is established among the experimented prediction models considering all techniques for the time horizon of 1 day, 3 days, 5 days, 7 days, 5 days and 30 days in advance, applying on all the datasets used in this study. For performance evaluation, statistical measures such as mean squared error (MSE) and correlation R are used. From the simulation result, it can be clearly suggested that BR outperforms other models for shortterm prediction especially for 3 days ahead. On the other hand, lM generates better prediction accuracy than BR- and SCGbased models for long-term prediction, especially for 7-day prediction. Keywords Artificial neural networks Autoregressive Bayesian regularization Deep learning Egyptian stock market Levenberg–Marquardt Stock price prediction
1 Introduction Financial projections and business predictions are of significant importance for capital investment industry, but due to several tangible and intangible factors, highly fluctuating trends are observed in stock markets, making the prediction a challenging task. These volatile economic and noneconomic factors including random exchange rates, changing & Essam H. Houssein [email protected] Mahmoud Dirar [email protected] Kashif Hussain [email protected] Waleed M. Mohamed [email protected] 1
Faculty of Computers and Information, Minia University, Minia, Egypt
2
Institute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
political situations, and inflation combine to put several stacks at risk. This demands efficient stock price prediction tools, in order for decision-makers to form the basis for profitable and well-informed investment decisions. Recently, stock price prediction has been an important research area where researchers have exploited the enticing benefits of machine learning methods [1], because the traditional statistical methods (regression, autoregressive integrated moving average, exponential average, etc.) often fail to produce accurate predictions when involved multiple nonlinear characteristics [2]. Moreover, these methods are limited to sta
Data Loading...
