0
$\begingroup$

I am a postgraduate student.

I would like to ask if there is a matter of multicollinearity, serious or not, when in a multiple linear regression the condition index in a dimension is below 15 (12.739) and for that dimension two variables have values greater than 0,9 (actually both have 0,99). Tolerance and VIF have acceptable values (for this specific variable TOL = O.937 and VIF = 1.O67) for this specific variable.

The model has three independent variables. The two other variables have acceptable values for valid interpretation (TOL, VIF, CONDITION INDEX AND VARIANCE PROPORTION)

I really thank you in advance.

$\endgroup$
2
  • 1
    $\begingroup$ This really needs more clarification and a better structure as it is pretty unreadable now. $\endgroup$ Sep 14, 2018 at 9:22
  • $\begingroup$ Welcome to Cross Validated. In order to maximise the chance that you get a good answer, you might need to clarify some of what you are asking for. Can you explain more fully what you mean by the following sentence: "the condition index in a dimension is below 15 (12.739) and for that dimension two variables have values greater than 0,9" $\endgroup$ Sep 14, 2018 at 9:26

1 Answer 1

1
$\begingroup$

Collinearity is not a "yes/no" type of question, it is one of degree. One way of assessing how serious it is is to use the perturb package in R (if you have R).

In general, I'd say shared variance that high indicates some sort of issue that should be investigated. What are these two variables? What is their correlation? Try a scatterplot of the two. (Correlation is not a good measure of collinearity because collinearity can involve more than two variables, but here only two seem to be involved).

Even if these two variables are not causing troublesome collinearity, it seems likely that they are not both adding information to the regression.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.