[Openspace] condition number (multicollinearity)
john.heppen at uwrf.edu
Tue Dec 4 11:05:13 CST 2007
A number less than 20 is ok. You maybe have to be a little more careful
in your interpretation and state the potential for interference from
German Muchnik Izon wrote:
> When I run an OLS regression in GeoDa I get a condition number of
> around 18. Is this acceptable in the empirical literature? I know that
> if it is 30 or above means that there is a serious multicollinearity
> problem but I want to make sure that a value of around 18 is acceptable.
> Thank you!
> Openspace mailing list
> Openspace at sal.uiuc.edu
More information about the Openspace