Linear Regression | What is Variance Inflation Factor ? How is it used?

Question

Use of VIF

in progress 0
TheDataMonk 55 years 2 Answers 1241 views Grand Master 0

Answers ( 2 )

  1. Avneet Singh
    0
    Reply
    Linear Regression | What is Variance Inflation Factor ? How is it used?

    The variance inflation factor is defined as shown in the picture. Where Rj^2 is the R-squared from the auxiliary regression, i.e., a regression of all features except jth variable on the jth variable. The Matrix shows how does the inclusion of the given feature inflate the variance of our model.

  2. swap007
    0
    Reply

    The Variance Inflation Factor (VIF) is the measure of how much the variance of
    the estimated regression coefficients are inflated as compared to when the
    predictor variables are not linearly related. It helps in detecting multicollinearity.

    VIF ranges from 1 upwards. A VIF of 1.5 tells you that the variance of a particular coefficient
    is 50% bigger than what you would expect if there was no correlation with other predictors.

    Generally, a VIF of 1 indicates no correlation
    a VIF between 1 and 5 indicates that the variables are moderately correlated
    a VIF of greater than 5 indicates high correlation between variables

Leave an answer

Browse
Browse