摘要
由于常规线性回归模型对个别异常数据敏感 ,所以回归方程欠稳定。建立了实测数据对回归直线的隶属度的计算公式 ,提出了以该隶属度为权重的模糊加权线性回归模型。该模型通过隶属度加权来削弱个别异常数据对回归直线的影响 ,从而达到提高回归方程稳定性的目的。
Owing to the conventional linear regression model is sensitive to abnormal data,therefore the regression equation is not stable enough.This paper established the calculation formula of subordinate degree of the measured data to regression straight line and put forward the fuzzy weighted linear regression model which takes this subordination degree as the weight.By means of weighted subordination degree this model weakens the influence of distinct abnormal data affecting upon the regression straight line and thus achieves the goal to enhancing the stability of regression equation.
出处
《机械设计》
CSCD
北大核心
2000年第10期11-12,29,共3页
Journal of Machine Design
基金
国家自然科学基金!资助项目
四川省跨世纪杰出青年学科带头人培养基金资助项目
牵引动力国家重点实验室开放课题基金!资助项目
机
