Papers by Keyword: Accumulated Generating

Abstract: Applying the reciprocal accumulated generating and the constructing background value of non-equidistant sequence of GRM(1,1) model based on fitting the first accumulated generating sequence with non-homogeneous exponential law for the problem of lower precision as well as lower adaptability in non-equidistant GM(1,1) model, the calculation formulas were deduced and a non-equidistant GRM(1,1) model generated by reciprocal accumulated was put forward. The grey GRM(1,1) model can be used in non-equidistant interval & equidistant interval time series and has the characteristic of high precision as well as high adaptability. Example validates the practicability and reliability of the proposed model.

Abstract: Monotonically decreasing sequence data for the traditional modeling method using the gray model accuracy is not high , and GM (1, 1) modeling method has inherent deviation , Model does not meet the compatibility condition, using Accumulated Generating Operation in reciprocal number ,make best use of last information and GM (1, 1) modeling , is deduced and the parameters optimization grey derivative calculation formula ,and then established GRM(1,1) based on accumulated generating operation in reciprocal number on the equidistance, Gray provides a new method of modeling . Data processing examples show that the model's practicality and reliability.

Abstract: The accuracy of the grey model is not high for the monotonic decreasing data and the error can not avoided. The model can not satisfy the compatibility conditions and then it is not accordant with the linear transformation. The paper adopts the equal interval unbiased model and takes one of the components in as the initial values. By this method it induces the parameters in the grey unbiased GRM(1,1) model and establishes the equal interval unbiased model based on the reciprocal AGO process. The model has high accuracy and has good value in the theory and practice. The examples show it is practical and reliable.