Modelagem do comportamento forward-looking dos índices setoriais no Brasil

• Ano de defesa: 2015
• Autor(a) principal: Sampaio, Glaylson Rodrigues
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Análise setorial; Preços de ativos
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/15433

The co-movements among asset prices suggest the action of exogenous influences, however, there is no consensus in the literature about which economic factors are responsible. This study investigates the sensitivity of the sector indexes BM&FBOVESPA to changes in expectations of market participants to the constant economic variables Focus Report. Make up the sample the seven sectoral indexes with complete series for the period January 2009 to June 2014, comprising 286 weekly observations. The approach used for the average modeling of indexes of series is based on multifactor pricing theory, using the forward-looking variables accompanied this report as candidates for factors. Given the presence of conditional heteroskedasticity in the sector indexes return series, it is proposed to GARCH framework for modeling the volatility. The results suggest that some expectations variables are significant for explanation of variations in Brazilian industry indices and the proposed multifactor models of returns can generate prediction gains evidenced by reduced of mean squared error of the return predictions.