• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.3.2019.tde-20032020-085902
Document
Author
Full name
Marcel Fernandes Dallaqua
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2019
Supervisor
Committee
Patah, Leandro Alves (President)
Faria, Ricardo Meirelles de
Rezende, Marcelo Novaes de
Title in Portuguese
Sistema de apoio à  decisão inteligente para Day Trading: contornando a racionalidade limitada e tomando boas decisões consistentemente com o uso da Inteligência Artificial.
Keywords in Portuguese
Inteligência artificial
Sistemas de apoio à decisão
Abstract in Portuguese
Para que possam tomar boas decisões consistentemente e alcançar seus objetivos financeiros, investidores devem enfrentar processos de tomada de decisão caóticos, complexos e frequentemente carregados de emoção. Nas últimas décadas, inúmeros sistemas de apoio à decisão vêm sendo propostos na literatura. Entretanto, a maioria dos estudos publicados deixa de fornecer regras para negociar no mercado e gerenciar os riscos, negligencia a avaliação do lucro dos métodos propostos e ignora a influência do comportamento humano na tomada de decisão e, consequentemente, nos resultados obtidos. Assim, verifica-se a necessidade de se explorar a seguinte questão: como contornar os limites da racionalidade para consistentemente tomar boas decisões e, consequentemente, obter resultados livres de vieses no mercado de capitais brasileiro? As próximas páginas verificam a hipótese de que, no day trading, os limites da racionalidade podem ser contornados com a aplicação de um sistema de apoio à decisão inteligente. Assim, objetivam medir a influência do comportamento humano no desempenho de um sistema de negociação financeira autônomo. Para isso, constroem e avaliam um sistema de apoio à decisão inteligente para day trading, empregando os métodos de simulação e experimento controlado no paradigma de pesquisa design science. Adicionalmente, propõem um framework para avaliação e aperfeiçoamento de sistemas de negociação financeira. E, por fim, demonstram ser possível contornar os limites da racionalidade, mas que, mesmo com um sistema autônomo, o impacto do comportamento no desempenho não é completamente nulo. Assim, contribuem, por um lado, diretamente com o processo metodológico para construção e avaliação de sistemas autônomos de negociação financeira, e, por outro, indiretamente com o desempenho inconsistente da grande maioria dos day traders, que acaba abandonando a atividade.
Title in English
Intelligent decision support system for day trading: getting around bounded rationality and consistently making good decisions with the use of artificial intelligence.
Keywords in English
Day trading
Design Science Research (DSR)
Intelligent Decision Support System (IDSS)
Abstract in English
To make good decisions consistently and finally achieve their financial goals, investors must face chaotic, complex and often emotionally charged decision-making processes. In the last decades, numerous decision support systems have been proposed in the literature. However, most of the published studies fail to provide rules for trading in the market and for managing risks, neglect profit evaluation for the proposed methods, and ignores the influence of human behavior in decision-making and, consequently, on the obtained results. Thus, it is found necessary to explore the following question: how to get around bounded rationality making good decisions consistently and achieving biases-free results in the Brazilian capital market? This thesis tests the hypothesis that, for day trading, it is possible to get around bounded rationality with the application of an intelligent decision support system. So, its goal is to measure the influence of human behavior on the performance of an autonomous financial trading system. Therefore, it builds and evaluates an intelligent decision support system for day trading, employing the methods of simulation and controlled experiment in the design-science research paradigm. Besides, a framework for evaluation and improvement of financial trading systems is also proposed. In the end, it shows that it is possible to get around the limits of rationality, but also that the impact of human behavior on the performance is not completely absent (even with the use of an autonomous trading system). Therefore, it contributes, on one hand, directly to que methodology for building and evaluating autonomous financial trading systems, and, on the other, indirectly to the inconsistent performance of the great majority of day traders, that eventually abandon the activity.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2020-05-29
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2021. All rights reserved.