Estratégia de Momento Aplicada ao Mercado de Ações no Brasil Utilizando Regressão Linear

• Ano de defesa: 2019
• Autor(a) principal: Santana, Samara Damascena de
• Orientador(a): Rodrigues, Carlos Alberto
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual de Feira de Santana
• Programa de Pós-Graduação: Mestrado em Computação Aplicada
• Departamento: DEPARTAMENTO DE TECNOLOGIA
• País: Brasil
• Palavras-chave em Português: Finanças Computacionais; Efeito Momento; Indicador k-ratio; Regressão Linear; Ibovespa
• Palavras-chave em Inglês: Computer finance; Momentum Effect; K-ratio Indicator; Linear Regression
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://tede2.uefs.br:8080/handle/tede/1341

The objective of the present work was to analyze the effects of the momentum strategies using linear regression in combination with the k-ratio indicator, when applied in the Brazilian stock market, verifying if their returns are better in relation to those obtained by Ibovespa in the same period. In addition to the central objective, it was also possible to describe which of the different approaches of the momentum strategy has the best result in Brazilian market, to verify the behavior of the strategy with the falling interest rate (Selic) and to evaluate momentum returns when compared to returns in Selic in a different period. To achieve the objectives, four algorithms of this strategy were developed and named: Traditional Moment, Moment S / A of volatility, Moment C / A ofvolatility, Momentwith k -ratio. These algorithms were applied to the subdivisions of a 650 stock data base and also across the data base. The base subdivisions were called: without volume criteria, smaller volumes, average volumes andlarger volumes, all in relation to the Ibovespa volume. These applications occur in two distinct periods: the period (2000-2018) and theperiod (2016-2018). The Mann-Whitney statistical test with 95 % confidence was used to analyze the significance of the results found. In the period (2000-2018), thereturns of momentum strategies were found when applied to the data base groups called average volumes and without volume criteria. In both bases, the moment with k-ratio was themost profitable approach, surpassingeventhe Ibovespa. In the period (2016-2018) with Selic in decline, the results were similar to the results obtained in (2000-2018). The highest return of the momentum strategies occurred on the bases without volume criteria and average volumes. Given the results meet, it wasfound that the momentum strategy has positive returns in the Brazilian market, whenapplied to stocks ofmedium volumes comparedto Ibovespa. Regarding the seactions, in general, the current strategies discussed here outperformed the market index, Ibovespa. Among thestrategiesthatsurpassedthe Ibovespa, the moment with k ratio was more profitable in longerperiodswith a stronger statistical significance than the others and the most profitable traditional moment in shorter periods, but with lower significance. Compared to Selic-based fixed income investments, returns on equity investments such as stocks are a profitable investment diversification option for the Brazilian investor for years to come, withthe prospect of low interest rates in the economy. The momentum effect continue to manifest and prove to be advantageous.