The Study of the Relationship between the Prediction Error and AGO Times in GM(1,1) Model

L.-H. Hsu (Taiwan)


numerical method, AGO times, GM(1,1) model, toolbox, Matlab and LabVIEW


The aim of this paper is by using the numerical method to design and to develop a toolbox, by which to analyze the relationship between the prediction error and AGO times in GM(1,1) model. Traditionally, the error analyses are all focused on the background value alpha, and most of GM(1,1) papers have only used AGO once. Although this method can decrease the value of error, we still feel that there are some insufficient exist. Hence, in this paper, we focus on the AGO times, and develop the toolbox, by which the Matlab and LabVIEW software platform can analyze the relationship between the AGO times and the prediction error. As the result, we can implement that the AGO times for GM(1,1) model are enough to be lower to once, and the prediction errors will be increased when AGO times are increasing as well.

Important Links:

Go Back