0

could anyone help me with my problem - I’m a beginner.

When doing multiple regression and checking for multicollinearity, I have found 2 VIF’s of 17, however, I’ve done some reading and few say it is okay to do multiple regression. Also, my Durbin-Watson is 1.927 which is alright.

Please let me know your thoughts, Thank you.

Moon Paris
  • 11
  • 1
    I have two answers on Cross Validated about the myths associated with feature correlation. (1) (2) – Dave Jun 07 '22 at 18:38
  • (1) VIF is not a great way to assess multicollinearity, despite its popularity. (2) Whether 17 is a problem depends also on how many variables and how much data you have, as well as which variables contribute to the near-collinearity and whether their estimates are of interest to you. These are some of the reasons additional information about your circumstances and goals would be welcome. – whuber Jun 07 '22 at 19:37
  • I am interested whether five factors of personality predict overall beliefs about revenge pornography? (Five variables measuring the five factors of personality (i.e., extraversion, agreeableness, conscientiousness, openness, and neuroticism).

    I have 500 participants. In regards to near-collinearity and estimates, I am completely lost about this because I am just a beginner. What I do know is that I have tried centralise means of the variables but it came with the same VIFs. For reference, 3 VIFs are around 1.5 while the other two are 16.7 and 17.1.

     – Moon Paris Jun 07 '22 at 19:56

0 Answers0