基于贝叶斯正则化的TDBPNN模型在中国外贸预报中的应用及评估

中国管理科学 ›› 2005, Vol. ›› Issue (1): 1-8.

• 论文 • 下一篇

基于贝叶斯正则化的TDBPNN模型在中国外贸预报中的应用及评估

祝树金, 赖明勇

湖南大学经济与贸易学院, 长沙, 410079

收稿日期:2004-03-30 修回日期:2004-12-03 出版日期:2005-02-28 发布日期:2012-03-07
基金资助:
国家自然科学基金资助项目(70273010);国家社会科学基金重点项目(04AJL006);博士点基金项目(200010532009)

An Application of the TDBPNN Model Based on Bayes’ Regularization to Forecasting China’s Foreign Trade and Evaluation

ZHU Shu-jin, LAI Ming-yong

College of Economics and Trade, Hunan University, Changsha 410079, China

Received:2004-03-30 Revised:2004-12-03 Online:2005-02-28 Published:2012-03-07

摘要/Abstract

摘要： 基于相空间重构的非线性预报思想,建立一个时滞的BP神经网络模型(TDBPNN),采用贝叶斯正则化方法提高BP网络的泛化能力,并将该模型应用于中国进出口贸易的预测,结果证明改进的TDBPNN模型具有较好的泛化能力,准确拟合了进出口贸易发展的历史值和趋势。区别于一般的预测评价,认为非线性预测不仅要注重数据拟合和精度改进,而且应该能够反映被预报系统的非线性特征。在分析模型预测精度的同时,通过计算拟合序列和原序列的非线性特征量进行模型评价,证实预测模型能够合理地“捕捉”到产生原序列的非线性系统的动力学特征。

关键词: 非线性预测, 进出口贸易, 相空间重构, BP网络, 贝叶斯正则化

Abstract: Based on nonlinear prediction ideas of reconstructing phase space,this paper presents a time delay BP neural network model,whose generalization is improved utilizing Bayes’ regularization.Furthermore the model is applied to forecast the import and export trades in China.The results show that the improved TDBPNN model has excellent generalization capabilities,which can not only learn the historical curve,but efficiently predict the trend of trade development.In contrast to conventional evaluation of forecasts,we assess the model by calculating the nonlinear characteristics of the predicted and original time series besides analyzing the precision of forecasting.The estimated values demonstrate that the dynamics of the system producing the original series has been reasonably captured in this model.

Key words: nonlinear prediction, import and export trades, phase space reconstruction, BP neural networks, Bayes’ regularization

中图分类号:

F201
C931

祝树金, 赖明勇. 基于贝叶斯正则化的TDBPNN模型在中国外贸预报中的应用及评估[J]. 中国管理科学, 2005, (1): 1-8.

ZHU Shu-jin, LAI Ming-yong. An Application of the TDBPNN Model Based on Bayes’ Regularization to Forecasting China’s Foreign Trade and Evaluation[J]. Chinese Journal of Management Science, 2005, (1): 1-8.

参考文献

[1] Lapedes R. and R. Farber. Nonlinear Signal Processing Using Neural Networks: Prediction and System Modeling[R], Teachnical Report LA-VR87-2662, Los Alamos National Laboratory, Los Alamos, New Mexico, 1987.
[2] Robert J. Van Eyden. The Application of Neural Networks in the Forecasting of Share Prices[M]. Finance and Technology Publishing, 1996.
[3] 魏巍贤,朱楚珠,蒋正华.基于人工神经网络的汇率预报[J].系统工程理论与实践,1996,(6):13-29.
[4] Tkacz G. Neural network forecasting of Canadian GDP growth[J]. International Journal of Forecasting,2001,17(1) :57-69.
[5] 吴德胜,梁樑.概率神经网络在财务预警实证中的应用[J].中国管理科学,2003,11(10):173-177.
[6] 沙磊,韩丽川.BP神经网络在上海市场需求预测中的应用[J].数学的实践与认识,2003,33(11):40-43.
[7] 许东,吴铮编著.基于Matlab6.X的系统分析与设计--神经网络[M].西安:西安电子科技大学出版社,2002.
[8] Foresee F. Dan and Hagan M. J. Gauss-Newton Approximation to Bayesian Learning [C]. Proceedings of the international Joint Conference on Neural Networks, June,1997.
[9] 王振全,徐山鹰.协整模型与预测--中国外贸出口的定量研究[J].国际贸易,2000(1):23-26.
[10] Takens F. Detecting Strange Attractors, In: Dynamical systems and turbulence (edited by D. A. Rand and L.S. Yang) [C]. Lecture Notes in Mathematics, vol. 898,Berlin: Springer.
[11] 王凌.智能化算法及其应用[M].北京:清华大学出版社,2001.
[12] MacKay, D. J. C. Bayesian Interpolation [J]. Neural Computation, 1992,4: 415-447.
[13] Grassberger P. and I. Procaccia. Measuring the Strang-eness of Strange Attractors[J]. Physica D, 1983,9:189-208.
[14] Principe,J. C.,A. RathieandJ. M. Kuo. Prediction of Chaotic Time Series with Neural Networks and the Issue of Dynamic Modeling[J]. International Journal of Bifurcation and Chaos,1992,2(4) :989-996.
[15] 吕金虎,陆君安,陈士华.混沌时间序列分析及其应用[M].武汉:武汉大学出版社,2002.

基于贝叶斯正则化的TDBPNN模型在中国外贸预报中的应用及评估

An Application of the TDBPNN Model Based on Bayes’ Regularization to Forecasting China’s Foreign Trade and Evaluation

PDF (PC)

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

Metrics

本文评价

推荐阅读 0