在对黄金市场进行分析时，通常根据黄金价格数据自身特点选取合适的模型进行建模预测，但因黄金价格数据本身的非线性特征比较明显，模型的选取往往较为困难，预测精度不高。利用神经网络的特性，建立了RBF神经网络，有效地解决了模型选择不当的难题。实证表明，RBF神经网络建立的非线性模型预测精度较高。
In the analysis of gold market ,suitable model is selected usually based on the characteristics of gold price data .However ,due to the evident nonlinear characteristic of the data itself ,model selection is often difficult and the prediction accuracy is doubted .RBF was established based on the characteristics of neural networks ,neural net-works,effectively solving the problem of improper model selection .Empirical study shows that RBF neural network based nonlinear model has better prediction accuracy .
参考文献
| [1] | 中国人民银行 .关于促进黄金市场发展的若干意见[EB/OL].http:∥finance.people.com.cn/GB/12333466.html,2014-03-05. |
| [2] | 史峰;王小川;郁磊.Matlab神经网络30个案例分析[M].北京:北京航空航天大学出版社,2010 |
| [3] | 范剑青;姚琦伟.非线性时间序列[M].北京:高等教育出版社,2005 |
| [4] | 张延利.基于 BP神经网络的黄金价格非线性预测[J].黄金,2013(07):8-9,10. |
| [5] | 陈勇,孔峰.BP神经网络预测全国私人汽车拥有量[J].计算技术与自动化,2003(03):67-70. |
| [6] | 魏巍贤.基于人工神经网络的汇率预报[J].系统工程理论与实践,1996(06):13-20. |
上一张
下一张
上一张
下一张
计量
- 下载量()
- 访问量()
文章评分
- 您的评分：
-
10%
-
20%
-
30%
-
40%
-
50%