小波分解在非平穩(wěn)時間序列預(yù)測中的應(yīng)用.pdf

1、上海交通大學(xué)碩士學(xué)位論文小波分解在非平穩(wěn)時間序列預(yù)測中的應(yīng)用姓名：鄧凱旭申請學(xué)位級別：碩士專業(yè)：計算數(shù)學(xué)指導(dǎo)教師：宋寶瑞200501015Applying Wavelet Decomposition in the Forecasting of Non- Stationary Time Series ABSTRCT Wavelet analysis is a great development in mat

2、hematics it overcomes Fourier’s limitation which cant’s analyze the signal in time and frequency at the same time and has the advantage of extracting characteristic of signal. In this thesis, we use this chara

3、cteristic of the wavelet to the forecasting of non- stationary time series. Before this, we often use statistical method to forecast the time series. But when the time series is non- stationary, the effect is

4、 not very good. We choose some proper wavelet function t o decomposing the series to a fix arrange. We use the wavelet decomposition to improve the station of the time series. And then, we use the statistic

5、al method to forecast the time series and get the result by reconstruction. We can see that the effect of the wavelet method is better then the statistical method because the station of the time seri