The Forecast of Price Index Based on Wavelet Neural Network

Financial time series are non-stationary, nonlinear, and stochastic, which makes prediction for them rather difficult. This article uses one method based on the wavelet analysis and the artificial intelligence to predict the A300 index in China and NASDAQ index in the USA. Comparing with wavelet-ARIMA model and simple BP neural network, our model(wavelet combined neural network) demonstrates superiority in predicting power. The results of different prediction lengths indicate that these methods are only suitable for short-term forecasts, their prediction for long-term is bad. The difference of forecasting between A300 and NASDAQ indicates that Chinese stock market is less efficient than that in the USA, the later may be weak efficiency.