Detecting multicollinearity from standard error?

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

LeonWalras
Posts: 17
Joined: Fri Mar 14, 2014 12:44 pm

Detecting multicollinearity from standard error?

Postby LeonWalras » Sat Aug 09, 2014 10:20 pm

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
Attachments
Regression question.JPG
Regression question.JPG (48.65 KiB) Viewed 2356 times

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Detecting multicollinearity from standard error?

Postby startz » Sat Aug 09, 2014 10:52 pm

The intuition is that the regression has very high explanatory power, but none of the individual coefficients is significant.


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 2 guests