Research On Improvement Of Smooth Degree Of Data Series And Grey Relation Analysis

The grey system theory has been greatly developed and applied in social economy system since it was put forward. However, because of the complexity of social economy activity, present grey systems theory has not fully satisfied the demand of actual problems. According to the defect of grey system theory, some improved methods in grey model and grey relation analysis are put forward on the basis of summarizing the present research achievements in this paper. Thus, the grey system theory can solve more actual problems and meet more needs of social system.In this paper, the smooth degree of original data series and its effect on model precision is studied and the more general method of improving smooth degree of data series is put forward. According to the general method, several kinds of new transformations are represented and the practical application shows the effectiveness and superiority of these methods. The defect of the GM(1,1) model on choice of initial value is analyzed theoretically, and a new method of GM(1,1) model based on optimum weighted combination with different initial is put forward. This method is applied into building a practical model and the results show its effectiveness. Then the data characteristic of non-equidistant GM(1,1) model is studied and two modeling methods are introduced. And the weighted non-equidistant GM(1,1) model is put forward. Its model mechanism is that the different data of non-equidistant series are given the different weighted values. Also a new method of grey relation analysis is proposed on the basis of summarizing various quantification...