As conclusion for this article, if the Wavelet transform is used for the return data, then there are no outlier, seasonal effects and other irregular effects. Generally the result of the approximation series under the wavelet transforms (rather by using Haar or Daubechies) is better than the original return data and more stable in variance, mean and no outliers. Furthermore, the forecasting using ARIMA (p, d, q) under the transformed series is better than forecasting directly, and also it gives more accurate results.