摘要
GDP数据序列分布往往具有非齐次指数性和非凸凹一致性的特点,运用传统的灰色模型预测GDP发展趋势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,可实现对呈不规则非齐次指数函数分布的数据序列的无偏拟合。运用无偏差GM(1,1,k)预测了河北省GDP,消除了GDP不规则分布对预测结果的影响,取得了满意的效果,平均预测误差为3.367 2%,比经典GM(1,1)的平均预测误差减小了47.035 8%。
The distribution of GDP data sequence often has the characteristics of non-homogeneous exponent and nonconvex-concave consistency. It is difficult to predict the development trend of GDP with the traditional gray model. The unbiased GM（1, 1, K） model is optimized from two angles of gray derivative and background value, which can realize the unbiased fitting of the data sequence with irregular uneven exponential function distribution. In this paper, the unbiased GM（1, 1, K） is used to predict the GDP of Hebei Province which the effect of the irregular distribution of GDP on the prediction results is eliminated, and the satisfactory results are obtained. The average prediction error is 3.3672%, and the average prediction error is reduced by 47.035 8% compared with the classical GM（1,1）.
作者
舒服华
SHU Fuhua(School of Mechanical and Electrical Engineering,Wuhan University of Technology,Wuhan,Hubei 430070,China)
出处
《衡水学院学报》
2018年第3期38-43,共6页
Journal of Hengshui University
