Page 1 of 1

Detecting multicollinearity from standard error?

Posted: Sat Aug 09, 2014 10:20 pm
by LeonWalras
Hello all,

A question I wanted to ask is that how do you determine if a regression equation exhibits multicollinearity from the standard errors displayed? I'm inclined to say this regression is multicollinear because of the high R^2, but I don't know how to reach this conclusion using the standard error?

The question is seen below from a previous year's past examination paper.

Thank you in advance,
Leon

Re: Detecting multicollinearity from standard error?

Posted: Sat Aug 09, 2014 10:52 pm
by startz
The intuition is that the regression has very high explanatory power, but none of the individual coefficients is significant.