区间灰数几何预测模型的参数优化  

Parameter Optimization of Interval Grey Number Geometry Prediction Model

作　　者：崔学海[1] 

机构地区：[1]重庆工商大学长江上游经济研究中心,重庆400067

出　　处：《统计与决策》2017年第11期10-14,共5页Statistics & Decision

基　　金：国家自然科学基金资助项目(71271226)

摘　　要：面积序列及坐标序列的模拟精度是影响区间灰数几何预测模型性能的重要因素,文章通过克莱姆法则建立面积序列与坐标序列的灰色模型参数无偏估计新方法,在此基础上构建了一种新的区间灰数预测模型;最后通过与传统的区间灰数预测模型模拟精度进行了比较,结果表明新模型具有更为优秀的模拟性能。The simulative precision of area sequence and coordinate sequence are important factors that exert impact on the performance of the geometric prediction model of interval gray number. This paper applies Cramer Rule to deduce a novel unbi- ased estimation method of area sequence and coordinate sequence, on the basis of which the paper establishes a new interval gray number prediction model. Finally, comparison is made between the accuracy of the proposed model and that of the traditional in- terval gray model without parameter optimization. The result shows that the novel model has better simulative performances.

关 键 词：灰色理论 预测模型 CRAMER法则 区间灰数 参数优化 

分 类 号：N941.5[自然科学总论—系统科学]