文章讨论的方法是在时序分析建模的基础上,运用了GM(1,1)来提取趋势项,然后对残差进行时序建模,形成灰色与时序的组合模型,在建模中的参数估计方法,采用抗差估计来取代常规的最小二乘估计法,来达到抵抗粗差的目的,从实例计算来看,效果比较好。
With a priori timing analysis modeling, the GM ( 1, 1 ) is used to extract the trend term, and then the timing modeling is implemented for the residuals, so as to realize the combination of time series analysis and gray model. For estimation of model parameters, the robust estimation method is used to replace the conventional least squares method. It can achieve the target of resistance to gross errors. The results of examples show that our new method is good.