A hybrid fuzzy time series model based on ANFIS and integrated nonlinear feature selection method for forecasting stock

Forecasting stock price is a hot issue for stock investors, dealers and brokers. However, it is difficult to find out the best time point to buy or sell stock, since many variables will affect the stock market, and stock dataset is time series data. Therefore many time series models have been proposed for forecasting stock price; furthermore the previous time series methods still have some problems. Hence, this paper proposes a novel ANFIS (Adaptive Neuro Fuzzy Inference System) time series model based on integrated nonlinear feature selection (INFS) method for stock forecasting. Firstly, this study proposed an integrated nonlinear feature selection method to select the important technical indicators objectively. Secondly, it used ANFIS to build time series model and test forecast performance, then utilized adaptive expectation model to strengthen the forecasting performance. In order to evaluate the performance of proposed model, the TAIEX and HSI stock market transaction data from 1998 to 2006 are collected as experimental dataset and compared with other models. The results show that the proposed method outperforms the listing models in accuracy, profit evaluation and statistical test.