Multicollinearity is the occurrence of high intercorrelations among two or more independent variables in a multiple regression model.This could be misleading and therefore wider confidence intervals that produces less reliable probability .
it is better to use different types of indicators rather than multiple indicators of the same type to avoid multicollinearity.
How to identity Multicollinearity
Two variables are considered perfectly collinear if their correlation coefficient is +/- 1.0.
*How can Multicollinearity be solve *
It can be done by calculating Tolerance(T) and Variance inflation factors(IVF).
If the T < 0.1 or VIF > 10 then that column must be drop to allow for better regression.
The link below can help calculate Multicollinearity on a set of independent variables datatab.net
Top comments (0)