Stock price prediction using DEEP learning algorithm and its comparison with machine learning algorithms

• Author: Gholamreza Mansourfar,Jamshid Bagherzadeh,Mahla Nikou
• Published date: 01 October 2019
• Date: 01 October 2019
• DOI: http://doi.org/10.1002/isaf.1459

RESEARCH ARTICLE

Stock price prediction using DEEP learning algorithm and its

comparison with machine learning algorithms

Mahla Nikou

1

|Gholamreza Mansourfar

1

|Jamshid Bagherzadeh

2

1

Faculty of Economics and Management,

Urmia University, Urmia, Iran

2

Faculty of Electrical and Computer

Engineering, Urmia University, Urmia, Iran

Correspondence

Gholamreza Mansourfar,Faculty of Economics

and Management, Urmia University. Urmia,

Iran.

Email: g.mansourfar@urmia.ac.ir

Summary

Security indices are the main tools for evaluation of the status of financial markets.

Moreover, a main part of the economy of any country is constituted of investment

in stock markets. Therefore, investors could maximize the return of investment if it

becomes possible to predict the future trend of stock market with appropriate

methods. The nonlinearity and nonstationarity of financial series make their predic-

tion complicated. This study seeks to evaluate the prediction power of machine‐

learning models in a stock market. The data used in this study include the daily close

price data of iShares MSCI United Kingdom exchange‐traded fund from January 2015

to June 2018. The prediction process is done through four models of machine‐

learning algorithms. The results indicate that the deep learning method is better in

prediction than the other methods, and the support vector regression method is in

the next rank with respect to neural network and random forest methods with less

error.

KEYWORDS

artificial neural network, deep learning, prediction, random forest, support vector regression

1|INTRODUCTION

Stock price indices are very significant among the world financial mar-

kets as the main criteria for evaluating the function of securities and

stock market. They are achieved from accumulation of stock price

movements of all companies or certain class of existing companies in

exchange market (Wang, Wang, Zhang, & Guo, 2012). Therefore, as

financial markets present a high degree of competition among partici-

pants (Parot, Michell, & Kristjanpoller, 2019), the study of stock price

index prediction models would be very necessary for investors to turn

the securities market to a profitable place. This leads to updating

investors' knowledge in making proper decisions in selection of an

appropriate portfolio, on the one hand, and the provision of interna-

tional investment opportunities for investors, on the other hand.

Nevertheless, prediction of financial markets is a complicated task,

in that financial time series are noisy, nonstationary, and irregular.

Although there are many statistical and computational methods for

prediction of these series (Atsalakis & Valavanis, 2009), it is still

identified as a difficult problem on financial variables and especially

price index. Predictions are not precise, but their rate of error depends

on the algorithm used. The important point is that, according to

researchers, the behaviour of most variables follows a nonlinear trend

in financial markets (Thomaidis, 2006); therefore, it is likely that linear

prediction does not yield appropriate results for examination of the

future path of financial variables.

In the financial literature, price prediction methods are classified

into four groups of technical analytical methods, fundamental analysis,

and prediction based on time series and machine learning. Through

discovering new patterns in historical data, machine learning intends

to identify the underlying function based on which data are formed

and realize the linear and nonlinear models that exist in these data

(Kalyvas, 2001). In past years, algorithms such as artificial neural net-

works and support vector machine (SVM) were widely used for predic-

tion of financial series and achievement of high precision in prediction

(Das & Padhy, 2012; Guo, Wang, Liu, & Yang, 2014; Lu, Lee, & Chiu,

2009). The results of studies indicate the considerable superiority of

Received: 22 February 2019 Revised: 20 September 2019 Accepted: 20 September 2019

DOI: 10.1002/isaf.1459

164 © 2019 John Wiley & Sons, Ltd. Intell Sys Acc Fin Mgmt. 2019;26:164–174.wileyonlinelibrary.com/journal/isaf