Beta Coefficients in the Brazilian Market Sectors. Fuzzy Regression vs Crisp Regression

ANTONIO TERCEÑO GÓMEZ; GLORIA BARBERÀ; HERNÁN VIGIER; YANINA LAUMANN

Girona

Congreso; XVIII SIGEF Congress; 2015

International Association for Fuzzy - Set

We consider that every decision-making process, and especially those using beta as a risk measure, is set in an uncertain environment. This works is a Continuation of the study of beta coefficients using fuzzy regressionanalysis, which begins in Terceño et al. (2011, 2014). Our objective is to use all the information provided by the market to determine the systematic risk. With the aim of incorporating every inaccuracy which accompaniesignorance about the future and the subjectivity associated with the decision making, we intend to advance in the calculation of the beta using the fuzzy regression model of Tanaka e Ishibuchi (1992) to calculate the sectorsbetas of the Brazilian Stock Market. The analysis with fuzzy regression can be applied with crisp data, uncertain or with a mixture of both. The aim of this work is, precisely, to compare the obtained results using the fuzzyregression with crisp data and uncertain data. After that, we make a comparison with the results to be obtained with the calculation of the beta by ordinary least squares. The comparison allows us to determine which of thesystems allows a better adaptation of reality.