The Structure Method and Application of Background Value in Grey System GM(1,1) Model (Ⅰ)
TAN Guan-jun (College of Mechanical and Electrical Engineering, Central South University of Technology, Changsha 410083)
In this paper, for the problem that grey GM(1,1) model often makes delay error for fitting high growth index series, we point out that the structure method of background value in grey GM(1,1) model has an important influence on the precision and adaptability of the model. From this point of view, the structure method of background value is researched, a simple calculating formula of background value is reestablished which has strong adaptability. An obvious feature of the new calculating formula of background value is that it makes GM(1,1) model have the ability to optimize the modeling results, the best fitting and forecasting precision can be obtained with it. The new formula is suited to building GM(1,1) model for both low growth index series and high growth index series. It is the key technology of enhancing GM(1,1) model precision and adaptability. The calculating precision of examples show this new formula is very effective.
【CateGory Index】： N941.5