基于小波变换的沪深300指数预测

Hushen 300 Index Price Forecasting Based on Wavelet Transform

摘要: 本文通过基于小波变换和未基于小波变换对沪深300指数日收盘价序列分别建立ARMA拟合模型并做短期预测，对其归一化均方误差(NMSE)进行比较，结果显示，由于小波变换良好的时频局域化特性，以及它的多分辨功能，使组合模型较之于单个预测模型对于沪深300指数的短期预测更优。

Abstract: This paper based on wavelet transform and non-wavelet transform established ARMA models to fit the daily closing price of Shanghai and Shenzhen 300 index and do a short-term forecast. It also compared their normalized mean square error (NMSE). Results display that due to the characte-ristics of wavelet transform which are good time frequency localization and its multi-resolution features, combined forecast model for short-term forecast of Shanghai and Shenzhen 300 index is superior to single forecast model.