Optimization and Application of initial Value of Non-equidistant New Information GM(1,1) Model
Aiming the problem of determining initial value of non-equidistant new information GM(1,1) model, researching modeling mechanism of non-equidistant new information GM(1,1) model which taking the newest component of original data as the initial value of response function of grey differential equation, the cause of the problem was found out, a new method for initiating value of non-equidistant new information GM(1,1) model was proposed to minimize the quadratic sum of its fitting error and the optimum formula of initiating value was detruded. The new non-equidistant new information GM(1,1) model with the proposed formula of initiating values has the characteristic of high precision as well as high adaptability. Examples validate the practicability and reliability of the proposed model.
Keywords: Initialization, Background value, GM(1,1),New information principle, Non- equidistant, Optimization, Grey system.
Download Full-Text
ABOUT THE AUTHOR
Youxin Luo
His current interests are information science, grey system, mechanics, and optimizing.
Youxin Luo
His current interests are information science, grey system, mechanics, and optimizing.