摘要
回归模型中自变量的高度相关关系导致了多重共线性的发生,传统的多重共线性检验方法不具有明确的检验标准,文章基于Farrar和Glauber提出的Bartlett统计量检验方法进行改进,解决了FG方法适用的普遍性,使该方法具有统计学上的显著意义。并且通过蒙特卡洛模拟验证了该方法的有效性。
The high correlation of independent variables in the regression model leads to the occurrence of multicollinearity,and the traditional multicollinearity test does not have a clear test standard. This paper is based on the Bartlett statistical magnitude test proposed by Farrar and Glauber to make improvement and obtain the universality of application of FG method, making this method statistically significant. Finally, the paper verifies the effectiveness of the proposed method by Monte Carlo simulation.
作者
朱钰
郑屹然
尹默
Zhu Yu;Zheng Yiran;Yin Mo(Institute of Statistics,Xi’an University of Finance and Economics,Xi’an 710100,China)
出处
《统计与决策》
CSSCI
北大核心
2020年第7期34-36,共3页
Statistics & Decision
