摘要
灰色建模要求的样本点少,不要求样本有较好的分布规律,而且计算量少,操作简便.而BP网络在对样本进行学习时,会对输出误差进行反馈校正,具有并行计算、分布式信息存储、容错能力强、自适应学习功能等优点.本文将灰色预测建模和神经网络技术融合起来建立灰色神经网络组合模型.应用于青海省宏观经济的预测,实证表明,该组合模型精度方面比常规灰色模型要好.组合的算法概念明确,计算简便,有较高的拟合和预测精度.它的提出拓宽了灰色模型的应用范围.
The foundation of grey model GM（1,1） need fewer samples that without better distributing order,but also do less calculating quantity.BP network adopts the back propagation when it learns samples,so it has ability of parallel calculation、distributed information memory、admitting-error,and so on.In this paper,we combine grey prediction method with the neural networks method to found the grey neural network Combinatorial model.This model is used for macroeconomics forecast in Qinghai Province.The practical example show that Combinatorial predicting model is definite in concept,convenient in calculation,good in fitting and precise in prediction,thus Combinatorial model improves the precision of the GM（1,1） model and enlarges its application scope.
出处
《青海师范大学学报（自然科学版）》
2011年第1期19-21,共3页
Journal of Qinghai Normal University(Natural Science Edition)
关键词
灰色系统模型
神经网络
组合预测模型
宏观经济预测.
grey system theory
neural network
combinatorial predicting model
redicting macroeconomics
