APPLICATION OF RECURRENT NEURAL NETWORKS IN THE PROVISION OF TIME SERIES: A STUDY OF STOCK EXCHANGE STOCK PRICES

Authors

DOI:

https://doi.org/10.12662/2359-618xregea.v11i2.p25-36.2022

Keywords:

deep learning, recurrent neural networks, price prediction, stock exchange

Abstract

Future events prediction in an assertive way has been the object of analysis by several researchers, especially in the financial area. In this context, there are countless possibilities for using this tool in the decision-making process of investment managers and analysts. This article aim to propose a recurrent neural network model based on the study of time series, oriented to the prediction and estimation of the price of shares on the Brazilian stock exchange. In this context, the present study enables the characterization of directions of financial trends and the prediction of prices from the training of the neural network, using real data from the Brazilian stock exchange in the year 2010. Regarding the methodology, the present research can be classified as applied, explanatory, in terms of its objectives, and quantitative in terms of its approach. The results obtained in this study reveal the ability to learn complex problems and, consequently, the possibility of application in other areas.

Downloads

Download data is not yet available.

Author Biographies

Gabriel Dilly Vieira Furtuoso, Engenheiro do Centro Tecnológico do Exército (CTEx).

Graduado em Engenharia de Computação pelo Instituto Militar de Engenharia (IME). Engenheiro do Centro Tecnológico do Exército (CTEx).

Marcos dos Santos, Instituto Militar de Engenharia (IME) e Centro de Análise de Sistemas Navais (CASNAV)

Doutor em Engenharia de Produção (pela UFF) na Linha de Pesquisa: Sistemas, Apoio à Decisão e Logística. Mestre em Engenharia de Produção (pela COPPE/UFRJ) na Linha de Pesquisa de Pesquisa Operacional. É Licenciado em Matemática e Especialista em Instrumentação Matemática (também pela UFF). Bacharel em Ciências Navais (pela Escola Naval - EN) com habilitação em Engenharia de Sistemas. Professor do Programa de Pós-graduação em Sistemas e Computação (PPgSC) do Instituto Militar de Engenharia (IME). Professor do MBA em Data Science e Analytics da Universidade de São Paulo (USP). Professor colaborador da UFF, atuando no Programa de Pós-graduação em Engenharia de Produção. Na Universidade Federal de Pelotas (UFPel), participa do projeto de pesquisa "Política Internacional e Gestão do Espaço Oceânico: o planejamento espacial marinho voltado aos estudos de defesa e segurança" onde desempenha a função de Coordenador da subárea "Segurança e Defesa".

Renato Santiago Quintal, Professor do Magistério Superior da Escola Naval (EN) e e Encarregado da Célula de Inovação Tecnológica daquela Escola (CIT-EN).

Doutor em Ambiente e Desenvolvimento pelo Programa de Pós-graduação em Ambiente e Desenvolvimento da Universidade do Vale do Taquari (PPGAD/Univates). Mestre em Ciências Contábeis (Faculdade de Administração e Finanças da Universidade do Estado do Rio de Janeiro/FAF-UERJ); Especialista em Educação Tecnológica (Centro Federal de Educação Tecnológica Celso Suckow da Fonseca/CEFET-RJ); Especialista em Comércio Exterior (Instituto de Economia da Universidade Federal do Rio de Janeiro/IE-UFRJ); Especialista em Gerenciamento de Projetos (Fundação Getúlio Vargas/FGV); Bacharel em Direito (Faculdade Nacional de Direito/UFRJ); Bacharel em Administração (Universidade Cândido Mendes/UCAM); Licenciado em Pedagogia (Universidade Federal do Estado do Rio de Janeiro/UNIRIO); e Bacharel em Ciências Navais, com Habilitação em Administração de Sistemas (Escola Naval/EN). Egresso dos seguintes cursos da Escola Superior de Guerra (ESG): Diplomacia de Defesa (CDIPLOD-2020), Direito Internacional dos Conflitos Armados (CDICA-2021), Governança em Defesa (CGED-2021), Análise de Crises Internacionais (CACI-2021) e Economia e Planejamento de Defesa (CEPD-2021). Professor do Magistério Superior e Encarregado da Célula de Inovação Tecnológica da Escola Naval (CIT-EN).

Downloads

Published

2022-05-03

How to Cite

FURTUOSO, Gabriel Dilly Vieira; SANTOS, Marcos dos; QUINTAL, Renato Santiago. APPLICATION OF RECURRENT NEURAL NETWORKS IN THE PROVISION OF TIME SERIES: A STUDY OF STOCK EXCHANGE STOCK PRICES. Journal Of Management Analysis, Fortaleza, v. 11, n. 2, p. 25–36, 2022. DOI: 10.12662/2359-618xregea.v11i2.p25-36.2022. Disponível em: https://periodicos.unichristus.edu.br/gestao/article/view/4006. Acesso em: 22 jul. 2024.

Issue

Vol. 11 No. 2 (2022)

Section

ARTICLES

License

AGREEMENT FOR THE CESSION OF AUTHOR’S RIGHTS

Through the present instrument, on the one hand, CENTRO UNIVERSITÁRIO CHRISTUS – UNICHRISTUS, henceforth referred to only as UNICHRISTUS, and, on the other,

1) Name _________________________________, ID number _______________________________,

2) Name _________________________________,ID number _______________________________,

3) Name _________________________________,ID number _______________________________,

4) Name _________________________________,ID number _______________________________,

5) Name _________________________________,ID number _______________________________,

henceforth referred to only as AUTHOR of the academic article whose title is ______________________________________, agree, in the best legal form, what follows:

ARTICLE FIRST – ON THE CESSION

The AUTHOR, their heirs and successors, in the terms of art. 49 of Law 9.610/98, cede the WORK for reproduction, publicizing, distribution, printing and publication by UNICHRISTUS, in any form or mean that exists or may come to exist.

Paragraph One. The aforementioned cession is non-exclusive and gratuitous, encompassing the totality of the WORK.

Paragraph Two. UNICHRISTUS may make the WORK available, in part or in total, for academic purposes, and is prohibited from altering its content, except for formal corrections that may be necessary.

Paragraph Three. The cession is valid in all countries, for versions in Portuguese or any translations to other languages, as UNICHRISTUS sees fit.

Paragraph Four. The original manuscript shall be delivered in the form indicated by UNICHRISTUS.

ARTICLE TWO – ON THE RESPONSIBILITIES

The AUTHOR declares that the WORK is theirs and is previously unpublished and is responsible for its contents and form, citations, references and other elements that compose it, having delivered it, when signing the present document, with the necessary language and methodology revision. As such, any judicial or extrajudicial measures relating to the content are of their entire responsibility.

Paragraph One. The AUTHOR declares there is no tacit prohibition attached to the publicizing of the WORK, and that it is free of any onus or impediments before Brazilian law.

Paragraph Two. Any burden relating to eviction belongs to the AUTHOR, who will also be responsible for compensating UNICHRISTUS, in case it is harmed by any judicial or extrajudicial means relating to the content of the work.

Paragraph Three. In case the work has multiple authors, they will be responsible in solidarity, except when proven otherwise.

ARTICLE THREE – ON JUDICIAL REPRESENTATION

The AUTHOR hereby constitutes UNICHRISTUS as their representative, thus authorizing it to act judicially or extra judicially on their behalf in cases relating to attacks against the WORK, be it by illegal reproduction, fraudulent edition or any form that constitutes an assault to intellectual property.

ARTICLE FOUR – FORUM

The parties hereby elect the forum of Fortaleza, in the state of Ceará, and renounce all others, as advantageous as they may be.

And because they deem the current agreement to be fair and just, the parties sign the present document in two copies.

Fortaleza, Ceará, on this ___ day of __________, 20__.

1) __________________________________________

2) __________________________________________

3) __________________________________________

4) __________________________________________

5) __________________________________________

Representing Centro Universitário Christus – UNICHRISTUS:

__________________________________________