0
$\begingroup$

I'm a beginner at statistics.

While doing the analysis, I did a simple regression and check out that some variables have low p-value. Therefore I thought they are influential variables for dependent variable.

So I did Multiple regression with influential variables, but p-values become not influentially low and less significant (less descriptive). Is this case some kind of multicolinearity? I've heard a lot about the opposite result (the more variable added, the insignificant variable becomes more descriptive), but I'm not sure about this case.

I would really appreciate it if you tell me why this happened and let me know whether or not I should take this variable in to multiple regression.

$\endgroup$
0
$\begingroup$

Without seeing that data - no one can tell you why this happens. Multicolinearity is indeed a possible cause. If these variable are continuous then you can check the VIF factor for each one of them. If indeed it caused by multicolinerity, then possible remedies is adding up variables, e.g., using principle components instead of the original explanatory variables or simply dropping the problematic variables.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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