Research on Optimum Weighted Combination GM(1,1) Model with Different Initial Value

In this paper, a new method of GM(1,1) model based on optimum weighted combination with different initial value is put forward. The new proposed model is comprised of weighted combination models with different initial value of raw data. Weighted coefficients of every model in the combination are derived from a method of minimizing error summation of square. The optimum weighted combination can express the principle of new information priority emphasized on in grey systems theory fully. The result of a numerical example indicates that optimum weighted combination GM (1,1) model presented in this paper can obtain a better prediction performance than that from the original GM(1,1) model.