I read nearly all topics about collinearity but still have some questions...

I know: multicollinearity is a problem because if two predictors measure approximately the same it is nearly impossible to distinguish them. They overlap each other. But WHY (??) is centering helpful for this(in interaction)? How can center to the mean reduces this effect?

  • 3
    $\begingroup$ I think there's some confusion here. Centering is not meant to reduce the degree of collinearity between two predictors - it's used to reduce the collinearity between the predictors and the interaction term. $\endgroup$
    – TPM
    May 2, 2018 at 14:34
  • $\begingroup$ Thank for your answer, i meant reduction between predictors and the interactionterm, sorry for my bad Englisch ;).. But the question is: why is centering helpfull? wat changes centering? i don't understand why center to the mean effects collinearity $\endgroup$ May 3, 2018 at 13:42
  • $\begingroup$ Please register &/or merge your accounts (you can find information on how to do this in the My Account section of our help center), then you will be able to edit & comment on your own question. $\endgroup$ May 3, 2018 at 13:45
  • $\begingroup$ I think you will find the information you need in the linked threads. Please read them. If it isn't what you want / you still have a question afterwards, come back here & edit your question to state what you learned & what you still need to know. Then we can provide the information you need without just duplicating material elsewhere that already didn't help you. $\endgroup$ May 3, 2018 at 13:47