Abstract: Stock price is the result of effects of a large number of factors The rule of price change is too complex to learn by neural network This paper uses wavelet packet theory to decompose price series into several subseries whose rule is relatively easy to learn by NN So the task of forecasting stock price is decomposed into forecasting in the decomposed series using NN The result of case study shows that the model of wavelet packet integrated neural network is more effect than same neural network model and wavelet integrated neural network model.

Key words: wavelet packet, neural network, stock market, forecasting