Title: Rough sets: technical computer intelligence applied to financial market

Authors: Paulo Henrique Kaupa; Renato José Sassi

Addresses: Industrial Engineering Post Graduation Program, Universidade Nove de Julho, Av. Francisco Matarazzo, Àgua Branca, Cep: 05001-100, São Paulo, Brazil ' Industrial Engineering Post Graduation Program, Universidade Nove de Julho, Av. Francisco Matarazzo, Àgua Branca, Cep: 05001-100, São Paulo, Brazil

Abstract: Investments in stock markets has called the attention of new investors by providing larger financial returns when compared to traditional investments, such as fixed income. However, this is a type of investment with a high degree of risk to which the investor must select a portfolio of stocks that combine maximised profit with minimised risk. Thus, correctly identifying the trends in stock prices with the help of a technique is critical for this investor. Computer intelligence techniques can be applied in this identification such as the rough sets theory. The rough sets theory was proposed as a mathematical model for knowledge representation and treatment of uncertainty, and it has been used subsequently in the development of techniques for classification in machine learning. The objective of this work was to apply rough sets in the selection of stocks for investment in the São Paulo Stock Exchange. The experiments were carried out with historical data extracted from the São Paulo Stock Exchange and the portfolio returns were compared with the Ibovespa Index, used as a benchmark. The results obtained positively point out to the application of rough sets in selecting stock portfolios for investment in the stock exchange.

Keywords: stock exchange; stock market investment; stock markets; rough sets; rough set theory; portfolio selection; stock portfolios; Ibovespa Index; financial markets; stock prices; price trends; mathematical modelling; uncertainty; Brazil.

DOI: 10.1504/IJBIR.2017.083268

International Journal of Business Innovation and Research, 2017 Vol.13 No.1, pp.130 - 145

Received: 21 Jan 2015
Accepted: 23 Mar 2015
Published online: 23 Mar 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article