If two variables are highly correlated, then they convey almost the same information. If one of these variables is used at a time, then it may not contribute significantly to the model. But simultaneous use of both the variables may result in significant contribution to the regression model. This kind of situation is known as multicollinearity. Critical level of correlation that may result in multicollinearity is considered to be 0.70/-0.70 for mathematical analysis.
Multicollinearity can be identified either by observing correlation coefficients or by using collinearity diagnostics such as Tolerance, VIF and Condition Index etc.
如果两个变量高度相关，那么他们将几乎相同的信息。如果这些变量使用一次，那么它可能不到模型有助于显着的。但两个变量同时使用可能会导致重大贡献的回归模型。这种情况被称为多重共线性。的相关性，可能会导致多重共线性的临界水平是数学分析0.70/-0.70。
多重共线性可确定通过观察相关系数或通过使用共线性诊断如宽容，VIF和条件指数等。
