摘要
灰色预测模型广泛应用于经济、医学、农业和水利等领域。用于同类型的预测模型还有很多,比如多元线性回归模型、指数平滑法、神经网络算法和TCE模型等。这些模型在计算方法和计算精度上有各自的优点。基于此,结合灰色预测模型、回归线性模型和指数平滑模型,得到精度更高、预测准确性更好的变权重组合预测模型,达到根据实际数据改变权重和快速解决实际问题的目的。以美国弗吉尼亚州药物数量预测为例,展示该模型的实际应用效果。
Grey prediction model is widely used in the fields of economy,medicine,agriculture and water conservancy.There are many prediction models for the same type,such as multiple linear regression model,exponential smoothing method,neural network algorithm and TCE model.These models have their own advantages in calculation methods and accuracy.Based on this,combined with grey forecasting model,regression linear model and exponential smoothing model,the variable weight combination forecasting model with higher accuracy and better forecasting accuracy is obtained,so as to change the weight according to actual data and solve practical problems quickly.Taking the forecasting of drug quantity in Virginia as an example,the practical application effect of the model is demonstrated.
引文
[1]傅立.灰色系统理论及其应用[M].北京:科学技术文献出版社,1992:10.
[2]张晓琴,陈佳佳,张振华.基于成分数据的变权重组合预测的权重确定[J].山西大学学报(自然科学版),2014,37(2):188-194.
[3]Zhang Q,Chen N,Zhao J.Combined Forecast Model of Refined Oil Demand Based on Grey Theory[J].Intelligence Information Processing&Trusted Computing International Symposium on,2010(10):145-148.
[4]田宜君,朱锋峰.基于GM(1,1)模型和灰色关联度的组合预测新方法[J].科学技术与工程,2009,9(4):842-844,851.