A New Grey Prediction IANGM (1,1, k, k²) Model

Abstract

In this paper, a new grey prediction IANGM (1,1, k , 2 k ) model with quadratic time-varying function is constructed, which is suitable for approximate homogeneous exponential type, approximate non-homogeneous exponential type, exponential linear combination type, and exponential parabolic combination type characteristic sequence. The undetermined coefficient method is used to determine the shadow equation matching the model, and the time response formula of the IANGM (1,1, k , 2 k ) model is derived based on the constant variation method. Through numerical simulation and double modeling analysis of China's domestic heat consumption and soft soil foundation settlement data from 2011 to 2020, the results show that the IANGM (1,1, k , 2 k ) model has higher simulation and prediction accuracy.