Estratégias de gestão de carteiras de investimentos no mercado brasileiro

• Ano de defesa: 2021
• Autor(a) principal: Miranda, Patrik Borges de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Curitiba; Brasil; Programa de Mestrado Profissional em Matemática em Rede Nacional; UTFPR
• Palavras-chave em Português: Finanças pessoais; Carteiras (Finanças) - Administração; Investimentos - Análise; Educação financeira; Bolsa de valores; Investimentos - Estimativas; Finance, Personal; Portfolio management; Investment analysis; Financial literacy; Stock exchanges; Investment - Estimates; CNPQ::CIENCIAS EXATAS E DA TERRA::MATEMATICA; Matemática
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/24662

We present and analyze the performance of investment strategies based on the classic Markowitz Mean Variance method and one of its variations, which we call Penalized Mean Variance, in the management of an investment portfolio in Brazilian companies shares listed on the São Paulo Stock Exchange. The analysis period includes the months affected by the economic crisis caused by the COVID-19 pandemic. The results indicate better performance of the Penalized Mean Variance strategy compared to Markowitz Mean Variance, as well as with the uniform allocation of capital among the portfolio’s assets, with the traditional 60-40 allocation and with the “Buy-and-Hold “. We simplified some procedures to make the strategies more applicable. Mean variance methods require an estimate of the expected return, for which we use the return historical average . We also analyzed the efficiency of the strategies in relation to the quality of these estimations. The application in the teaching of financial education was exemplified considering portfolios with few assets, using quadratic functions, matrices, vectors and statistics.