A novel stock price forecasting method using the dynamic neural network

With the rapid development of economic and people's investment requirements, stock has been an important part in modern society. In order to forecast the stock market, the most important issue is to analyze the stock market investment. Influence factors of the stock market include four aspects, that is, (1) Domestic economic and policy, (2) International environment, 3) Stock features, and 4) Technical judgement. To forecast stock price, we construct a dynamic neural network (named as NARX) model. NARX neural network means the nonlinear autoregressive with exogenous input, and NARX neural network is able to analyze the structure of nonlinear systems and time series based system. Finally, we choose two stocks to make performance evaluation, which are 1) Poly Real Estate (600048) and 2) Industrial and Commercial Bank of China (601398). Experimental results demonstrate that the proposed algorithm can forecast stock price with high accuracy.