nÃo hà / O cenÃrio de incerteza no mercado de capitais brasileiro, durante o perÃodo eleitoral, contribuiu para a formaÃÃo de clusters de volatilidade, que tornou difÃcil a precificaÃÃo dos ativos negociados na bolsa brasileira. Neste contexto, esta pesquisa visa investigar se este cenÃrio de incerteza poderia ter acarretado na quebra de algum dos maiores bancos brasileiros com capital aberto. As aÃÃes dos bancos analisados foram: Banco do Brasil ON, Bradesco PN e ItaÃUnibanco PN. A MÃtrica para essa pesquisa foi a de Value-at-Risk (VaR). Dois desses modelos sÃo ditos incondicionais no que se refere à variÃncia: o VaR Gaussiano incondicional e o VaR Best Fitting incondicional. Os outros dois modelos sÃo chamados de condicionais, assumindo que a volatilidade varia ao longo do tempo. A mÃtrica que melhor representou as sÃries temporais atravÃs do Backtesting foi o VaR Logistic condicional. A partir da estimaÃÃo do melhor modelo, identificou-se que os bancos nÃo quebrariam, mas poderiam ter acumulado expressiva desvalorizaÃÃo. / The scenario of uncertainty in Brazilian capital market, during the election period, contributed to the formation of volatility clusters, that made it difficult to pricing of the securities traded on the Brazilian stock exchange. In this context, this research aims to investigate if this scenario of uncertainty could have led to the breakdown of some of the largest Brazilian banks with open market. The actions of the banks analyzed were: Bank of Brazil ON, Bradesco PN and ItaÃUnibanco PN. The metric for this research was to Value-at-Risk (VaR). Two models that are said unconditional as regards the variance: unconditional Gaussian VaR and VaR Best Fitting unconditionally. The other two are called conditional models, assuming that the volatility varies over time. The best metric that represented the time series by Backtesting was conditional Logistic VaR. From the estimation of the best model, it was identified that the banks would not break, but could have accumulated significant devaluation.

No description available.