The stock market is a stochastic, dynamic environment and is in constant evolution, and its prediction represents a big challenge. Many studies presented in the state of the art are facing this challenge, by making use of Artificial Neural Networks (ANN) as a tool to make such prediction. In this paper a comparative study is made with different methods in order to predict the Brazilian stock market through the Bovespa Index. An ANN was developed and its performance was compared against a hybrid model, in which a Genetic Algorithm (GA) is proposed as an alternative to improve the performance of this ANN. The results obtained were an average accuracy of 55.04% and 55.73% respectively, demonstrating that algorithms such as a GA have the capability of improving the performance of ANN for the stock market prediciton.