Forecasting Stock Market Using Wavelet Transforms and Neural Networks and ARIMA (Case study of price index of Tehran Stock Exchange)

The goal of this research is to predict total stock market index of Tehran Stock Exchange, using the compound method of ARIMA and neural network in order for the active participations of finance market as well as macro decision makers to be able to predict trend of the market. First, the series of price index was decomposed by wavelet transform, then the smoothʹs series predicted by using the ARIMA and low-frequency parts of the signal was predicted by using neural network method then this predicted was compound with inverse wavelet transform. The main attention of this paper is investors and traders to achieve a method for predict stock market. Concerning the results of previous researches which confirm the relative superiority of non-linear models in price index prediction, an appropriate model has been offered in this research by compounding the non-linear method and linear method such as neural network and ARIMA with using wavelet transform, The results indicate superiority of the designed system in predicting price index of Tehran Stock Exchange. This paper by compounding the linear and non-linear method issues pattern to predict stock market, to encourage further investigation by academics and practitioners in the field.