In this paper introduces the optimization of ensemble neural networks with fuzzy integration type-1 and type-2 for application of the prediction of complex time series, the methods used for optimization are Genetic Algorithms (GA) and Particle Swarm Optimization (PSO) for optimization of ensemble neural networks for integration of network response is made with typo-1 and type-2 fuzzy systems. The time series that considered in this paper to compare the Hybrid Approach With Traditional Methods is the Taiwan Stock Exchange (TAIEX), and results shown are for the optimization of the structure of the ensemble neural network with the type 1 and type 2 fuzzy integration. Simulation results show that the ensemble approach produces good prediction of the Taiwan Stock Exchange.