Forecasting a Stock Trend Using Genetic Algorithm and Random Forest

Rebecca Abraham, Mahmoud El Samad, Amer M. Bakhach, Hani El-Chaarani, Ahmad Sardouk, Sam El Nemar and Dalia Jaber
Additional contact information
Rebecca Abraham: Huizenga College of Business, Nova Southeastern University-SBE, 3301 College Avenue, Fort Lauderdale, FL 33319, USA
Mahmoud El Samad: School of Arts and Sciences, Lebanese International University, Mouseitbah, Mazara P.O. Box 146404, Lebanon
Amer M. Bakhach: School of Arts and Sciences, Lebanese International University, Mouseitbah, Mazara P.O. Box 146404, Lebanon
Hani El-Chaarani: College of Business Administration, Tripoli Campus, Beirut Arab University, Beirut P.O. Box 11-50-20, Lebanon
Ahmad Sardouk: Faculty of Economics and Business Administration, Tripoli Campus, Lebanese University (UL), Beirut P.O. Box 6573/14, Lebanon
Sam El Nemar: Faculty of Business Administration, AZM University, Tripoli P.O. Box 1010, Lebanon
Dalia Jaber: School of Arts and Sciences, Lebanese International University, Mouseitbah, Mazara P.O. Box 146404, Lebanon

JRFM, 2022, vol. 15, issue 5, 1-18

Abstract: This paper addresses the problem of forecasting daily stock trends. The key consideration is to predict whether a given stock will close on uptrend tomorrow with reference to today’s closing price. We propose a forecasting model that comprises a features selection model, based on the Genetic Algorithm (GA), and Random Forest (RF) classifier. In our study, we consider four international stock indices that follow the concept of distributed lag analysis. We adopted a genetic algorithm approach to select a set of helpful features among these lags’ indices. Subsequently, we employed the Random Forest classifier, to unveil hidden relationships between stock indices and a particular stock’s trend. We tested our model by using it to predict the trends of 15 stocks. Experiments showed that our forecasting model had 80% accuracy, significantly outperforming the dummy forecast. The S&P 500 was the most useful stock index, whereas the CAC40 was the least useful in the prediction of daily stock trends. This study provides evidence of the usefulness of employing international stock indices to predict stock trends.

Keywords: computational or mathematical finance; stock trend prediction; random forest; genetic algorithm; features selection (search for similar items in EconPapers)
JEL-codes: C E F2 F3 G (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1911-8074/15/5/188/pdf (application/pdf)
https://www.mdpi.com/1911-8074/15/5/188/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jjrfmx:v:15:y:2022:i:5:p:188-:d:796978

Access Statistics for this article

JRFM is currently edited by Ms. Chelthy Cheng

More articles in JRFM from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().