다중공선성을 무시해도 되는 순간이 있다?

 

 

https://ukchanoh.wordpress.com/2015/02/16/multicollinearity/

언제 다중공선성(multicollinearity)을 무시해도 괜찮은가?

 

1. 관심변수가 아닌 통제변수인 경우

  - 관심변수의 VIF 값은 낮은 경우

 

2. interaction term 또는 power와의 관계

  - 증명될 수 있다고 함

 

3. 3개 이상의 범주를 가지는 더미변수

 

 

 

 

https://statisticsbyjim.com/regression/multicollinearity-in-regression-analysis/

Multicollinearity in Regression Analysis: Problems, Detection, and Solutions - Statistics By Jim

1. Multicollinearity affects the coefficients and p-values, but it does not influence the predictions, precision of the predictions, and the goodness-of-fit statistics. If your primary goal is to make predictions, and you don’t need to understand the role of each independent variable, you don’t need to reduce severe multicollinearity.
   → 예측이 목적이며, 설명변수에 대한 이해가 필요하지 않다면 다중공선성을 그리 걱정하지 않아도 된다??
     ㆍThe fact that some or all predictor variables are correlated among themselves does not, in general, inhibit
        our ability to obtain a good fit nor does it tend to affect inferences about mean responses or predictions of
        new observations.  —Applied Linear Statistical Models, p289, 4th Edition.