What does multicollinearity mean? This is a problem of collinearity between two or more explanatory variables, which can be quickly detected using EViews. The following article explains to you the method of using EViews to test multicollinearity. If you need to learn, please hurry up and learn with the editor.
Simple correlation coefficient test:
By testing the correlation coefficient r between x1
After entering the data, proceed as shown in the figure:
It can be seen that r=0.785534 between x2 and x3, which is very close to 1, so there is collinearity between x2 and x3, that is, there is a multicollinearity problem.
Comprehensive statistical test:
Usually, if the R^2 and F values are both large and the t value is small, it indicates the existence of multicollinearity.
The steps are the same as OLS inspection. They all use Quick--Estimate Equation and enter Y C X1 X2 X3 in the dialog box.
As can be seen from the above table, there is a high degree of linear correlation between the explanatory variables. Although the overall linear regression fit of the equation is good, the parameter t value of the X1 X2 X3 X7 variable is not significant, and the sign of the X3 X6 coefficient is contrary to economic significance. It shows that the model does have serious multicollinearity.
The above explains the specific operation process of using EViews to test multicollinearity. I hope friends in need can learn it.
