摘要
实际应用指数平滑法时,由于主要靠经验选取平滑参数与初值,预测结果往往不合理.通过对传统预测模型的分析,提出了扩大平滑参数选取区间以及初值选取模型化的思路,得到了可以克服预测滞后的平滑参数选取区间,获取了以所有历史数据按指数平滑规律加权后的初值.并用该优化方法检验了ERP项目实施中的生产经营目标预测值,取得了满意的效果.
In practice, people choose coefficients of exponential smoothing mainly depending on their experiences, therefore, the predicted results were not reasonable. After analyzing the traditional predict model, this paper has put forward a new method to expand the range of choosing of coefficient of exponential smoothing and to model the choosing of initial value. The method obtained the range of choosing of coefficient of exponential smoothing which can overcome the hysteresis of prediction and the initial value after assigning the historical data ratio according to the law of method of exponential smoothing. Moreover, this new method was used to predict and verify the objective valuation of production operations in implementing of ERP project to get the satisfactory result.
出处
《江南大学学报（自然科学版）》
CAS
2005年第3期316-319,共4页
Joural of Jiangnan University (Natural Science Edition)
关键词
指数平滑法
平滑参数
初值
优化
method of exponential smoothing
coefficient of exponential smoothing
initial value
optimize
