Approximate non-homogeneous index sequence GM(1,1) model of grey derivative optimization.

Based on the original sequence for non-homogeneous exponential of the grey derivative of GM(1,1) model, this paper uses forward difference quotient and backward difference quotient of the weighted average as the grey derivative of the GM(1,1) model, and gives the concrete expression of the weighted coefficient λ to establish the new optimized grey derivative of GM(1,1) model, which adapts to the non-homogenous exponential sequence. This paper proves that this model possesses white exponential law coincidence, gives a method to solve parameters and gets expression of parameters. The examples show this model has higher precision and can completely fit for strict non-homogeneous index series. [ABSTRACT FROM AUTHOR]