首页   按字顺浏览 期刊浏览 卷期浏览 Combining Neural Network Forecasts on Wavelet-transformed Time Series
Combining Neural Network Forecasts on Wavelet-transformed Time Series

 

作者: ALEX AUSSEM,   FIONN MURTAGH,  

 

期刊: Connection Science  (Taylor Available online 1997)
卷期: Volume 9, issue 1  

页码: 113-122

 

ISSN:0954-0091

 

年代: 1997

 

DOI:10.1080/095400997116766

 

出版商: Taylor & Francis Group

 

关键词: Keywords: Dynamical Recurrent Neural Networks;Time-series Prediction;Wavelet Transform

 

数据来源: Taylor

 

摘要:

We discuss a simple strategy aimed at improving neural network prediction accuracy, based on the combination of predictions at varying resolution levels of the domain under investigation (here: time series). First, a wavelet transform is used to decompose the time series into varying scales of temporal resolution. The latter provides a sensible decomposition of the data so that the underlying temporal structures of the original time series become more tractable. Then, a dynamical recurrent neural netork is trained on each resolution scale with the temporal-recurrent backpropagation algorithm. By virtue of its internal dynamic, this general class of dynamic connections network approximates the underlying law governing each resolution level by a system of non-linear difference equations. The individual wavelet scale forecasts are afterwards recombined to form the current estimate. The predictive ability of this strategy is assessed with the sunspot series.

 

点击下载:  PDF (181KB)



返 回