This objective of this research is to compare the forecasting models of S&P 500 index with 3 models- ARIMA, ARIMA with GARCH and ARIMA with E-GARCH. The secondary data are used to predict daily the values of S&P 500 since January, 1 to October, 31 2014. The performance of forecasting models in term of accuracy is measured by using of Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). The most appropriate model of S&P 500 is ARIMA with E-GARCH which given the minimal MAPE and RMSE.