一种改进不等间距灰色预测模型

传统不等间距灰色模型UGM(1,1)及其改进型都是基于指数模型建立的,仅对指数变化规律序列有较好的预测精度,而对于常见的线性变化序列则预测误差较大。针对该问题,通过模型拓展,在现有文献模型的基础上增加线性因素,并采用新陈代谢的思想,提出一种改进不等间距灰色预测模型AUGM(1,1),并进行实例仿真比较分析。结果表明:改进模型在预测精度和实用性上均有较大改善,且克服了传统灰色预测模型不适用于线性变化序列预测的局限,拓宽了灰色预测模型的适用范围。

An Amendatory Unequal Interval Grey Prediction Model

GM(1,1) model is widely applied in the uncertainty prediction of less data and poor information, but it is often limited in true application because it is established with equal interval, then the grey prediction model with unequal interval performs a more important realistic function. Currently, the traditional unequal interval grey model (UGM(1,1)) and its amendatory model are established based on exponential model, which could not fit on linear series prediction. In order to overcome the drawbacks of the traditional unequal interval grey model (UGM(1,1)) and its amendatory model, a new amendatory unequal interval grey model (AUGM(1,1)) is proposed by adding linear ingredient and adopting the metabolism theory. The result of simulation and the comparative analysis show that the amendatory model is better than the UGM(1,1) model in accuracy and practicality, and overcomes the shortage that the traditional one could not be used in linear series, in this way, the application field of the grey model is expanded.