基于小波分析的金融时间序列消噪方法及应用

Wavelet Based Denoising Method and its Application in Financial Time Series

利用小波在信号消噪方面的应用,结合时间序列的预测模型,提出了一种基于小波分解的消噪预测模型,并将其与原预测模型进行比较.最后的比较试验结果发现,应用消噪后的数据进行模型预测比原始数据进行直接预测相对误差更小,从而精度更高.

Based on wavelet decomposition and its application in signal denoising,and combined with time series model for prediction,this paper presents a prediction model aiming at attenuating the noise level.By comparing the proposed prediction model with the original prediction model,the results indicate that the proposed prediction model fed with denoised data gives smaller relative errors and consequently higher precision.