Algoritmo para otimização da estratégia de investimento em desempenho contrário

Detalhes bibliográficos
Ano de defesa: 2021
Autor(a) principal: Eduardo de Abreu Moraes
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade Federal de Minas Gerais
Brasil
Programa de Pós-Graduação em Administração
UFMG
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: http://hdl.handle.net/1843/44112
Resumo: The contrarian investing strategy consists of identifying a movement of a majority of investors in the stock market and trading against that movement. Although there is literature on the subject confirming the effectiveness of the strategy in generating excess return in relation to the market index, there seems to be no scientific work whose objective was to present the best way to implement the strategy in a given country . The present work aimed to fill this gap by estimating how each strategy implementation parameter contributes, on average, in the generation of Information Ratio. These estimates were made using the Ordinary Least Squares Method, performed on a database of simulation results (10,000 for the Brazilian case and 1,200 for the US case) of different random implementations of the strategy. These simulations were performed using an algorithm that represents the implementation of the strategy. Based on the coefficients estimated by the OLS, it was possible to identify the combination of parameters that maximizes the Information Ratio. The out-of-sample results for the Brazilian case were in the sense that this strategy optimization procedure is capable of generating an Information Ratio greater than the average of the random simulations, but still negative. In the US case, the results were inconclusive due to the small number of simulations performed due to the high computational cost, related to time and processing capacity.