摘要
将小波分析与BP神经网络相结合,构建了股指期货价格预测模型。选取沪深300股指期货从上市日至2013年8月20日的收盘价格数据作为样本,运用sym8小波变换对数据进行降噪处理,分别运用降噪前后的数据对BP神经网络进行训练和检验。结果表明,降噪数据可以有效提高股指期货价格预测的效果。
A forecasting model for the stock index future price was built with wavelet analysis and BP neural network. Choosing the CSI 300 index futures closed price series from the beginning to August 20, 2013 as sample, the sym8 wavelet transform was used to filter noise of the data. The BP neural network was trained and tested with the original data and the denoised date respectively. Results show that the denoised data can significantly increase the effectiveness of forecasting for the stock index future price.
出处
《青岛大学学报(自然科学版)》
CAS
2014年第1期101-105,共5页
Journal of Qingdao University(Natural Science Edition)
关键词
小波降噪
BP神经网络
股指期货
价格预测
wavelet-basecl denoising
BP neural network
stock index future
price forecasting