摘要
由于上海市社会消费品零售额数据序列分布呈非齐次指数的特点,传统的灰色预测模型难以获得理想的效果。无偏差GM (1,1,k)模型从灰导数和背景值两个角度优化,实现了对于非齐次指数函数的无偏拟合。文中运用无偏差GM (1,1,k)预测上海市社会消费品零售额,取得了满意的效果。平均预测误差为1.404 55%,比传统的GM (1,1,k)模型的降低了73.629 7%,比经典的GM(1,1)模型的降低了40.866 7%。
Due to the distribution of retail sales of social consumer goods in Shanghai shows a nonhomogeneous index,the traditional grey forecasting model is difficult to achieve ideal results.The unbiased GM（1,1,k）model is optimized from two aspects of grey derivative and background value,and the unbiased fitting of non-homogeneous exponential function is realized.In this paper,we use the GM（1,1,K）to predict the retail sales of social consumer goods in Shanghai,and have achieved satisfactory results.The average prediction error is 1.404 55%,which is 73.629 7%lower than the traditional GM（1,1,k）model and 40.8667%lower than the classical GM（1,1）mode.
作者
舒服华
宋良美
SHU Fuhua;SONG Liangmei(School of Continuing Education,Wuhan University of Technology,Wuhan 430070,China)
出处
《上海商学院学报》
2018年第4期15-21,共7页
Business Economic Review
