SHAZAM Community
Ask Your Question

What can cause multicollinearity in OLS?

asked 2011-04-06 13:52:00 +0000

A SHAZAM User asked:

I am a student at XXX University and am using the Shazam program to create a hedonic pricing model of laptops for a fourth year course.

My problem is that I have used dummy variables to signify the makers of each laptop. When I try to perform the OLS function, I get a message saying "Matrix Inversion failed in row 19". " This usually indicates perfect multicollinearity with variable constant".

I have searched the internet to no avail for an answer. When I run the regression without the dummy variables, the OLS works. How can I perform OLS on my data including the dummy variables? Do I have to specify that they are dummy variables somewhere in the code?

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2011-04-06 13:53:45 +0000

David gravatar image

updated 2012-05-30 12:22:34 +0000

Try using the NOCONSTANT option on the OLS command to remove any constant.

Also make sure you have N-1 dummy variables.

edit flag offensive delete link more
Login/Signup to Answer

Question Tools


Asked: 2011-04-06 13:52:00 +0000

Seen: 9,368 times

Last updated: May 30 '12